|
|
| Elliott Sound Products | Low-Power DC Supplies |
 Main Index
Articles Index
|
As an extension from Small Power Supplies (Part I) and Small Power Supplies (Part II), this article concentrates on practical solutions, without being sidetracked by the many extra details provided in the first two articles. There is some duplication, but not as much as you might think when looking at them. The first article has more detail about regulators and how they work, but that's a purely theoretical examination that won't help you to build a supply. The second looks at transient response and noise, which are largely irrelevant to the supply ideas described here.
When building projects, there are countless reasons that you'll need a low-voltage power supply to power 'stuff' that has little or nothing to do with the audio. These range from 12V trigger circuits (so an external 12V input turns the gear on), to power a soft-start such as Project 39, or to provide power to a speaker protection board (e.g. Project 33). Your project may require a push-on/ push-off circuit such as described in Project 166, or use a PIC or microcontroller that needs its own power, either full-time or just when the amp (or whatever else) is powered on. The question is which supply is the best?
There's no simple answer, as some auxiliary circuitry may only need a few milliamps, others might need a great deal more. If it's permanently on, reliability (and safety against possible fire) becomes an issue you have to consider, and you ideally need something that will last for at least the life of the final product, and preferably more. This may mean a simple mains transformer-based PSU if you don't need much current, and although they draw more power in standby than a modern switchmode supply, they tend to have an indefinite life (20+ years is usually easily achieved).
Across the Web, there are countless designs for low current (typically 1A or less) power supplies for preamps, small PIC based projects, ADCs, DACs and almost any other project you can think of. Many are very basic, using nothing more than a resistor and zener diode for regulation, while others are very elaborate. For most beginners and many experienced people alike, it often becomes harder than it should be. You have to make a decision, based on what you need (voltage and current), how much you're willing to spend, and expected life. If you need to power a relay (or several), consider that a 'typical' 10A relay with a 12V coil has a resistance of ~270Ω, and will draw 44mA, a dissipation of a bit over ½W. Higher voltage coils draw less current, but for a given size of relay, the power is fairly constant regardless of the voltage rating.
Ultimately, the final choice depends on the application, but for most ancillary gear, a 12V, 1A supply will cover most requirements. If you're using a conventional (i.e. 50/60Hz transformer, rectifier, filter and regulator), you need a transformer of around 30VA to get a clean 12V, 1A regulated supply. Where the current needed is low (~100mA), a 6VA transformer will suffice. Sometimes you might not even need that much, so you may get away with a 2-3VA tranny. You need to beware of the pitfalls (see Section 1.1 which looks at the ratings in more detail). A small SMPS (switchmode power supply) will often be more economical, but you sacrifice long-term reliability.
There may be applications where a 5V supply is preferred, perhaps for equipment that has Bluetooth or LAN connections that need to remain active. If that same supply is expected to activate relays, you need much higher current for a 5V relay than a 12V relay. For example, where a 12V relay may draw ~45mA, one of the same 'family' with a 5V coil will draw almost 110mA. The power consumed is the same though, so using a 5V supply certainly isn't out of the question. However, some ancillary equipment may not be able to function with 5V - P33 and P39 for example. The voltage is too low for the circuits to operate normally. One solution is to use a 12V supply with a secondary regulator to provide 5V. Small switchmode buck converters (step down) can be used to get high efficiency. However, the no-load current of these may be higher than a simple linear regulator.
There is an endless fascination by some to build the smallest and cheapest power supply possible. Many circuits can be found that don't even use a transformer, and while some have acceptable or at least 'adequate' warnings about safety, others do not. Transformerless power supplies are not considered here (see Transformerless Power Supplies; How To Configure Them Properly) for more info. In general, these are discouraged, because they are inherently dangerous.
All of the designs shown are intended for use where the DC is fully isolated from mains voltages. Make sure that you read the Dangerous Or Safe? - Plug-Packs (aka 'Wall Warts') Examined article before you embark on the use of an AC/DC switchmode supply. Some are likely to be lethal (especially if purchased from eBay, Amazon or Ali Express (for example). Many of these claim to be approved, but some are incapable of passing the most rudimentary approvals tests. The majority of the linked article is not included here, but it is very important that you understand that some SMPS are complete rubbish and/ or dangerous.
If you are not experienced with mains wiring, do not attempt the following circuits. In some countries it may be unlawful to work on mains powered equipment unless you are qualified to do so. Be aware that if someone is killed or injured as a result of faulty work you may have done, you may be held legally responsible, so make sure you understand the following ...
WARNING : The following description is for circuitry, some of which requires connection to mains voltage. Extreme care is required to ensure that the final installation will be safe under all foreseeable circumstances (however unlikely they may seem). The mains and low voltage sections must be fully isolated from each other, observing required creepage and clearance distances. All mains circuitry operates at the full mains potential, and must be insulated accordingly. Do not work on the power supply while power is applied, as death or serious injury may result.
For anyone who is unfamiliar with the terms 'creepage' and 'clearance' as applied to electrical equipment, they are defined as follows ...
Creepage: The shortest distance across a surface (PCB fibreglass or other insulating material) between conducting materials (PCB traces, etc.). Allow at least 8mm for general purpose equipment.
Clearance: The shortest distance through air between conductors. Again, 8mm is recommended, but it may be reduced if there is an insulation barrier between the conductors.
All countries have electrical wiring codes and standards, and compliance may be voluntary, implied or (in a few countries) mandatory (at least for some products). In any case, if a product is found to be dangerous, there will usually be a recall, which may be mandatory if the safety breach is found to be a built-in 'feature' of the product that renders it unsafe or dangerous. It is the responsibility of anyone who builds mains powered equipment to ensure that it meets the requirements that apply in the country where it's built or sold. The authorities worldwide take electrical safety seriously, and woe betide anyone who falls foul of the standards (and subsequently the law courts) by killing or injuring someone.
The power supplies described here are intended to power 'ancillary' circuitry, such as a speaker protection circuit, or perhaps a microcontroller or a motorised volume control. For powering preamps and other audio circuits, you'd typically use the P05 power supply, which is designed specifically for powering audio circuitry. I've shown 12V as the output voltage for the examples, but it can range from 5V up to 24V, depending on your needs.
The general schemes shown here range from around 50mA up to 1A. Lower current means lower cost, so there's no need to build (or obtain) a 1A supply if you only need 50mA, unless the cost is low enough to justify the added current capability. This is especially important for linear supplies, where the transformer is the most costly item. A 2VA transformer may be obtained for less than AU$10, and can supply up to 70mA. A 7VA transformer (250mA DC) will be about 50% more, but if you need more (say 18VA for 600mA DC) you'll pay an extra 25% again. Above that, the prices are significantly higher from most suppliers. As a general rule, assume that the maximum DC output current is roughly half the AC output current. The rectification and smoothing process has a poor power factor, and 2:1 is a safe margin (albeit generous in some cases). The current ratings listed above assume a transformer with a 15V secondary. The DC output will be around 20V with light loading, sufficient to allow for a regulator.
The theory of small supplies depends on their technology (linear vs. SMPS), but all we're interested in doing is obtaining a power supply that can be incorporated in a chassis as part of the main circuitry. This may be used to power a Project 33 speaker protection, a Project 39 inrush limiter, or any of the other things that you may wish to include in your construction. You can use a basic regulator from the main power supply, but that may not be advised for any number of reasons. In particular, the dissipation of the regulator may become excessive, especially of the main supply voltage is greater than 35V or so.
However, it remains an option and is included here because it's often the easiest way to get a low-voltage supply with the minimum of fuss. In general, this approach has limited current (around 100mA maximum) because the regulator may be dead simple, but it will dissipate power and will need a heatsink. This instantly increases the cost unless the chassis is aluminium (at least 1.5mm thick) that can be used as the heatsink. Note that circuits such as the Project 39 inrush limiter should never be operated from the main supply. If there's a fault, the circuit gets no power, and damage is guaranteed.
Since most hi-if products are powered from the mains, we need to galvanically isolate the output of the supply from the mains voltage. This is a vital safety requirement, and cannot - ever - be ignored, regardless of output voltage or power requirements. Galvanic isolation simply means that there is no electrical connection between the mains and the powered device. A transformer satisfies this requirement, but is not the only solution. One could use a lamp and a stack of photo-voltaic (solar) cells, but this is extremely inefficient. However, this technique is used in some MOSFET isolated gate driver ICs, but they only have to output a few microamps. Because most of the alternatives are inefficient or just plain silly, transformer based supplies represent well over 99.99% of all power isolation methods. Switchmode supplies also use a transformer, so they are included.
Transformers only work with AC, so the output voltage must be rectified and filtered to obtain DC. This is shown in Figure 1.1 - the transformer, rectifier and filter are shown on the left. For simplicity, single supply circuits will be examined in this article - dual supplies essentially duplicate the filtering and regulation with the opposite polarity. Since the idea here is to power ancillary circuitry, a negative supply is rarely needed. The filter is the first stage of the process of ripple (and noise) removal, and deserves some attention. However, many applications aren't particularly fussy, and while the next circuit can be improved, in many cases there's simply no point.
C1 (the filter capacitor) needs to be chosen to maintain the DC (with superimposed AC as shown in Figure 1.2) above the minimum input voltage for the regulator. If the voltage falls below this minimum because of excess ripple, low mains input voltage or higher current, noise will appear on the output - even if the regulator circuit is ideal. No conventional regulator can function when the input voltage is equal to or less than the expected output. It can be done with some switching regulators, but that is outside the scope of this article. Remember that the transformer's output current will be roughly twice the DC current. The regulation of small transformers is generally awful, so the simple circuit shown in Fig 1.1 is only suitable for around 150mA DC output, requiring a transformer with no less than a 4VA rating. The secondary voltage is 15V because small transformers have very poor regulation. You might get away with a 12V secondary, but there's very little headroom.
In the above schematic, there is about 300mV RMS (950mV peak-peak) ripple at the regulator's input, but only 10mV RMS (34mV p-p) at the output of the discrete regulator. This is a reduction of 30dB - not wonderful, but not bad for such a simple circuit. Load current is 120mA. With the addition of 1 extra resistor and capacitor to create a filter going to the base of Q1, ripple can be reduced to almost nothing. If you wish to experiment, replace R1 with 2 x 560Ω resistors in series, and connect the junction between the two to ground via a 100µF capacitor. This will reduce ripple to less than 300µV - 62dB reduction. Alternatively, one might imagine that just adding another large cap at the output would be just as good or perhaps even better. Not so, because of the low output impedance. Adding a 1,000µF cap across the load reduces the output ripple to 3.8mV - not much of a reduction.
While simple, a discrete regulator will actually cost more to build and use more PCB real estate than a typical 3-terminal IC regulator. The IC will also outperform it in all significant respects. You must also remember that the discrete regulator has no current limiting, so a shorted output will almost certainly destroy the transistor! It's not difficult to add basic current limiting, but even in its simplest form it will add a low-value resistor and a transistor or a couple of diodes. On the positive side, the discrete regulator with a bigger transistor can handle much a higher input voltage. If you use a Darlington (e.g. TIP122 as used in the Rev-B P33 circuit), the input voltage can be up to 100V. R1 would be increased to suit, sized to provide a nominal base current of 5mA ...
R = ( Vin - Vout ) / 5m For example, for a 56V supply ...
R = ( 56 - 12 ) / 5m = 8.8kΩ (use 8.2kΩ, preferably at least 0.5W)
The formula is 'close enough'. Aiming for accuracy is not required (and would be pointless) because there are too many variables. You'll need to use a 13V zener diode (or two series diodes) to compensate for the extra base-emitter junction of the Darlington transistor.
The discrete regulator in Figure 1.1 is very basic - it has been simplified to such an extent that it is easy to understand, but it still works well enough for many basic applications. The output ripple of the IC version is not shown, but will generally be well below 1mV p-p. Prior to the introduction of low-cost IC regulators, the Fig. 1.1 circuit used to be quite common, and a very similar circuit was common using valves (vacuum tubes). Early voltage references were usually neon tubes, designed for a stable voltage. These will not be covered in this article.
Referring to Fig. 1.2, it should be obvious that the filter capacitor C1 removes much of the AC component of the rectified DC, so it must have a small impedance at 100Hz (or 120Hz). If the impedance is small at 100Hz, then it is a great deal smaller at 1kHz, and smaller still at 10kHz (and so on). Ultimately, the impedance is limited by the ESR (equivalent series resistance) of the filter cap, which might be around 0.1Ω at 20°C. Using a larger capacitance reduces the ripple, but doesn't change the average DC voltage. If C1 is changed to 2mF (2,000μF), the input (and output) ripple is halved.
It is important that capacitive reactance is not confused with ESR. A 1,000µF 25V capacitor has a reactance of 1.59Ω at 100Hz, or 15.9Ω at 10Hz. This is the normal impedance introduced by a capacitor in any circuit, and has nothing to do with the ESR. At 100kHz, the same cap has a reactance of only 1.59nΩ (nano-ohms), but ESR (and ESL - equivalent series inductance) will never allow this to be measured. The ESR will typically be less than 0.1Ω, and is generally measured at 100kHz. Indeed, at very high frequencies, the ESL becomes dominant, but this does not mean that the capacitor is incapable of acting as a filter. It's effectiveness is reduced, but it still functions just fine. Some people like to add 100nF caps in parallel with electros, but at anything below medium frequency RF (less than 1MHz), such a small value of capacitance will have little or no effect. While this is easily measured in a working circuit, few people have bothered and the myth continues that electrolytic caps can't work well at high frequencies.
Contrary to popular belief in some quarters, electrolytic capacitors do not generally have a high ESL. Axial caps are the worst simply because the leads are further apart. ESL for a typical radial lead electro with 12mm lead spacing might be expected to be around 6nH. A short length of track can make this a great deal worse - this is not a fault with the capacitor, but with the PCB designer.
Unfortunately, a simple linear circuit as shown above needs a transformer, the cost of which is often greater than the cost of a complete switchmode AC/DC converter. It might be possible to find one for no more than (say) AU$10.00 or so (a 1.9VA transformer may be as little as AU$8.00), but the transformer size (in VA) needs to be twice the product of DC voltage and current, before regulation. A 15V, 2VA transformer can deliver 133A AC, but expecting more than 60mA DC is most unwise. In general, my recommendation would be a maximum DC output current of no more than 60mA. These small transformers have terrible load regulation, so the output voltage collapses quickly when the output is loaded.
Always consider the highest input voltage allowable for a regulator IC. For the common 7805/ 12/ 15 devices, that's 35V absolute maximum, or 40V for the 7824. For adjustable regulators (LM317/ 337) they quote the input-output differential (40V), so in theory you could have an input of 50V and an output of 12V (38V differential). However, at some point the regulator will die (if the output is shorted or during startup when it has to charge a capacitor). I strongly recommend that the maximum input voltage should be no more than 35V!
One way of reducing the voltage is to use zener diodes in series with the input. If the supply is 50V, a pair of 12V zeners in series will allow up to 50mA (the current for a 12V, 1W zener is 83mA - at full power. This can't be sustained and the zener(s) will overheat and fail. A better approach is to use the discrete circuit in Fig. 1.1, with a higher power series-pass transistor (Q1), and set for an output voltage of ~20-24V for a 12V output. If you were to use the suggested TIP122 (65W) Darlington transistor, with a good heatsink you could easily draw up to 1A (over 30W transistor dissipation!). Of course, if this is intermittent it's not a problem. This arrangement is far better than any alternatives, and while the transistor and zener will cost more than the regulator, at least the survival of the latter is assured. This same circuit can be used with the switchmode buck regulator described in Section 4. The pre-regulator's output voltage should be higher to reduce dissipation - The LM2596 can handle an input of up to 40V.
If you wanted to use a purely discrete supply, the one shown above should meet your needs. It includes (very basic) current limiting that will prevent destruction if the output is shorted. If the voltage across R3 exceeds 0.7V (about 350mA), Q2 conducts and (proportionally) removes base current from Q1, limiting the current to a nominal 350mA. In reality, it will be somewhat more into a short circuit. With the 50V supply shown, Q1 will dissipate close to 20W with a shorted output, so a heatsink is essential. The maximum allowable output current with a 50V input supply is about 800mA, which is just inside the SOA curve for the TIP122. At an output current of around 200mA, Q1 will dissipate 8W. That's rather a lot of heat to dispose of on a continuous basis.
Because of the extra filter (R1, R2 and C2), ripple rejection is about 60dB at 120mA output. This is more than enough for ancillary equipment, and adds almost no cost to the design. Overall, this is a fairly convenient solution, but it isn't suited to 'permanently on' equipment because it relies on the main supply for its operation. While it could be used with P33 (for example) the new PCB already has an on-board regulator (albeit simplified - similar to Fig. 1.1 discrete).
While this topic might seem very simple, judging from the number of emails I get asking about it, perhaps it's not so simple after all. Most people don't give transformer selection a second thought, which may be because the specifications are provided in the project itself, or because it seems to be so easy that you can't go wrong. Well, you can go wrong, and end up with unexpected results. Of these, the most common is the final voltage after rectification. With small transformers at light loading, the voltage will often be much higher than expected, and when loaded, lower than expected.
Transformer selection depends on many factors. The desired output voltage and current determine the transformer size, but the relationships are more complex than they may seem at first. Small transformers (< 10VA) have poor regulation because they have a high winding resistance. That means that you almost always need a higher voltage than you thought, so for a 5V output you'd generally need to start with a nominal output voltage of at least 7.5V AC. In theory, your unregulated output will be around 10V, but with no load it will be more than that. At full load (DC) it will be less than 10V, and you may not even have enough 'headroom' to ensure regulation without ripple breakthrough.
I tested a 5VA, 18V (9+9V) transformer, capable of 277mA AC output at full (resistive) load. With no load and 230V mains, the output was 21V RMS, and it was just under 18V with a 65Ω load. The primary resistance measured 707Ω, with 6.62Ω for the secondary (a total equivalent series resistance of about 12.5Ω). When connected to the Fig. 1.1 bridge and filter cap, the average DC output is 28.7V with no load, falling to 21V DC (average, 20.5V minimum) with a 170mA load. The AC output current measured 276mA RMS - close enough to the maximum allowed (277mA RMS). Everything changes if the transformer has a higher or lower VA rating! There are many 'simple' formulae suggested online, and they give simple answers that are almost always simply wrong!
For a 5V regulated supply using the transformer I tested, the two secondaries would be in parallel, rather than series. The unloaded DC voltage will be ~11V, falling to ~9.5V (average, 8.5V minimum) with an output current of 340mA. Because we expect a higher current, C1 would be increased to 2,000μF (2 x 1mF in parallel). The minimum voltage (the troughs of the ripple waveform) is increased to 9.1V, and the total secondary current is 546mA (just within the transformer's VA rating). Note that the voltages are not simply halved, because the diode voltage drop becomes more significant at lower voltages.
This is the reason that I published the series of articles on transformers. Nothing is as straightforward as it seems (and is so often presented). At low voltages, you almost always need a higher nominal output voltage than you might guess. This is doubly true if the voltage is to be regulated, and despite claims you may see that LDO regulators are the answer, they often cause other issues (such as unexplained oscillation) that can be difficult to solve. See the Low Dropout (LDO) Regulators article for more.
Remember that the mains voltage can increase/ decrease around the nominal value (230V or 120V), often by 10% or more. If you only have 10% headroom for a regulator, you will get ripple on the output if the mains voltage falls, and you'll also get higher regulator dissipation if the mains voltage rises. These anomalies must be accounted for in a design. Sometimes it doesn't matter though, because the voltage supply for many auxiliary circuits isn't at all critical. If relays have to be activated, the voltage must be above the minimum allowable voltage. For a 12V relay, that's generally about 9V, but it varies (always consult the datasheet!).
A regulator (in almost any form other than a zener diode) is an amplifier. Admittedly the amplifier is 'unipolar', in that it is designed for one polarity, and can only source current to the load. Very few regulators can sink current from the load, but shunt regulators are an exception! Since amplifiers can oscillate, it follows that regulators (being amplifiers) can also oscillate. As the bandwidth of a regulator is increased to make it faster, it will suffer from the same problems as any other wide bandwidth amplifier, including the likelihood of oscillation if bypassing isn't applied properly.
The regulator itself has two primary functions. The first is to provide a stable output voltage, and the second is reduction of the power supply filter noise - mainly ripple, and this pretty much comes free when the voltage is regulated. The regulated voltage may not be especially accurate, but this is rarely an issue.
The output impedance should be low, because this allows the voltage to remain constant as the load current changes. For example, if the output impedance were 1Ω, then a 1A current change would cause the output voltage to change by 1V. This may not be an issue with some circuits, but it will be unacceptable for others. One might normally expect the output impedance to be less than 0.1Ω, and that's easily achieved - even with simple designs.
In order to maintain low impedance at very high frequencies, an output capacitor is almost always required. This will be in addition to any RF bypass capacitors that are required to prevent oscillation. A 10μF output cap is usually quite sufficient to ensure stability. The output capacitor generally has little affect on the output ripple or noise, but it can help to provide instantaneous output current for nonlinear loads.
Remember that in any real circuit, there will be PCB traces that introduce inductance. Capacitors and their leads also have inductance, and it is theoretically possible to create a circuit that may act as an RF oscillator if your component selection is too far off the mark (or your PCB power traces or wiring are excessively long). In the common applications that are covered by this article, wiring inductance will never be a problem.
Bypassing is especially important where a circuit draws short-term impulse currents. This type of current waveform is common in mixed signal applications (analogue and digital), and the impulse current noise can cause havoc with circuitry - an improperly designed supply path can cause supply glitches that cause false logic states to be generated. Even the ground plane may be affected, and great care is needed in the layout and selection of bypass caps to ensure that the circuit will perform properly and not have excessive digital noise. Again, this is unlikely to be an issue for common ancillary circuits.
Maximum power dissipation, maximum current and internal protection are all things that need to be considered. These are dependent on the type of regulator, and the specifications and terminology can vary widely. Many of the parameters are far too complex to provide a simple 'figure of merit', and graphs are shown to indicate the transient performance (load and line) and other information as may be required to select the right part for a given task.
One special family of regulators are called LDO (low drop-out) regulators. Where a common regulator IC might need 2 to 5V input/output differential, an LDO type will generally function down to as little as perhaps 0.6V between the input and output. These are commonly used in battery operated equipment to maximise battery life. Some of these devices also have very low quiescent current, so there is a minimum of power wasted in the regulator itself. They are covered in Low Dropout (LDO) Regulators, but in general it's better to use a 'standard' regulator unless you really need the low-dropout. They make no sense for mains powered supplies.
Very few (especially non-audio) applications really need anything more than the traditional fixed voltage regulators, such as the 7812, 7815 and 7824 (positive) and 7912 etc. (negative). They are not ultra-quiet (electrically) at up to 90μV (15V version), but the noise is generally (but not always) immaterial when the circuit is only used for ancillary circuitry. Their ripple rejection is at least 54dB with an input-output differential of 10V. They include current limiting and over-temperature protection. Output current is ≥1A.
A 7812 (or 7912) has a typical output range of from 11.5V to 12.5V, so expecting the voltage to be exact is unrealistic and unnecessary. The load regulation (i.e. the change in output when the load current is changed) is anything from 12mV to 150mV when the load current is changed from 5mA to 1.5A. For this test, the input voltage is maintained constant. The dropout voltage is 2V, so the input voltage (including ripple) must be 2V higher than the output voltage at all times. See Fig. 1.2 (red trace) to see how ripple is measured. The minimum voltage in the graph is about 15.7V.
Ripple rejection is quoted as a minimum of 54dB to a typical value of 74dB, somewhat dependent on the input voltage headroom (at least 5V is a good idea if possible). These figures can be bettered by using the LM317/337 variable regulators. They have lower noise and better ripple rejection than the much older fixed regulators, but in most circuits it makes no difference whatsoever. Of more importance is the fact that they are variable, so you can keep a few on hand to regulate to any voltage you need (within their maximum input voltage range).
There are quite a few other regulator types on the market, but the National Semiconductor types seem to have the lion's share of the market as far as normal retail outlets are concerned. Not that there is anything wrong with them - they perform well at a reasonable price, and have a very good track record for reliability. While one can obtain more esoteric devices (with some searching), many of the traditional manufacturers are concentrating on switching regulators, and don't seem to be very interested in developing new analogue designs (other than LDO regulators).
Switchmode regulators are also available as a single IC, but they need more (and more expensive) support components. Of course they are also more efficient, so heatsink requirements are usually minimal for a few hundred milliamps output. The design process for many of these ICs is daunting, especially for most hobbyists. If you select a switchmode controller IC with an external switching MOSFET you gain a wider range of input voltages (up to 100V or more), but the ICs are almost all SMD, and the design process becomes much harder. An example is the LTC3894, with up to 150V input and adjustable output voltage (0.8 to 60V). However, it's SMD and not inexpensive, and there are many external parts needed (at least 8 capacitors, 6 resistors, a P-Channel MOSFET and an inductor).
While there are many discrete or semi-discrete linear regulators to be found in various books, websites (including this site) and elsewhere, they are usually only ever used because no readily available IC version exists. An example is the ESP P96 phantom power regulator - this design is optimised for low noise and the relatively high voltage needed by the 48V phantom system. Regulation is secondary, since the phantom power voltage specification is quite broad. It is still quite credible in this respect, but it has fairly poor transient response, which is not an issue for the application.
Many people would consider a switchmode buck (step down) regulator to be the easiest way to get (say) 12V from a 40-70V main supply rail. While this is true up to a point, most of the ICs you'll find are only rated for a maximum input voltage of around 30-40V, but often less. One IC that I've used is the LM2596T-ADJ, and it's surprisingly easy to get it working provided you're not after the highest possible current and efficiency.
The circuit is taken from Project 220, and it's a well tried circuit. The maximum input voltage is 40V, and the output can be adjusted from 1.23V up to 37V (the latter assumes a 40V supply). With a maximum output current of 3A, it can do most things you need. The datasheet provides very comprehensive formulae for determining the inductor value, but for around 200mA or so a 100μH inductor is generally fine. For higher current, the inductance needs to be lower, with thicker wire and a core that will not saturate. If that happens, bad things quickly follow.
These are available from various on-line 'auction' sites as a complete module, for little more than you'd expect to pay for the IC. So, while it's dead easy to build one on Veroboard (and I've done so), it will almost certainly cost more than a pre-built module. As noted above, there are other devices, with some even including the inductor in the package (e.g. WPMDH1200601/ 171020601). These are not cheap ICs though - expect to pay almost AU$30 for the IC alone. This is not viable for most hobby applications, and it's probably marginal for commercial designs as well.
You can built a very basic switchmode buck converter with nothing more than a cheap CMOS IC, a suitable MOSFET and an inductor (plus resistors and capacitors of course). The viability of this approach depends on your application, but in most cases it's simply not worth the effort. While I'm all for experimentation, if you're installing a circuit as part of an amplifier (for example) it's better to stay with something simple that can be repaired or replaced if (when?) it fails. SMPS are more prone to failure than simple linear circuits, and will almost always be harder to repair (especially when the IC becomes obsolete).
If you need to accommodate a supply voltage above 40V, you can use the Fig. 1.1 discrete circuit to supply the IC. You lose efficiency (and Q1 may require a heatsink), but it's a low-cost option that will work well. As the allowable input voltage of switchmode ICs increases, so does their cost. Most are also far more complex than the one shown, meaning that there are more things to go wrong.
Shunt regulators have some advantages over traditional series regulators, despite their low efficiency and comparatively high power dissipation. It's uncommon to see shunt regulation used any more, but they are useful at low current or where some ripple can be tolerated. The advantages of shunt regulators are that they are inherently short-circuit proof, can sink current from the load as well as sourcing current to the load and they provide (almost) fool-proof over voltage protection, including transient suppression.
Naturally, there are also disadvantages, as is to be expected. They have comparatively high power dissipation regardless of load current, and simple versions may have relatively poor overall performance. However, they are still worth considering where the load current is low (e.g. 10-20mA or so).
The simplest shunt regulator consists of nothing more than a resistor and a zener. If designed properly, this is a very simple power supply arrangement, and offers acceptable performance for many low-current applications. They are very rarely used where the circuit needs more than around 100mA or so, because dissipation becomes a real problem. Consider a shunt regulator expected to supply 12V at 100mA, fed from a 42V amplifier supply. In an 'ideal' world, the feed resistor will dissipate 3W continuously, regardless of load current! In reality it will be at least 5W to allow for voltage variations from the main supply.
This is one of the reasons that there are very few shunt regulators used in modern equipment. This is not necessarily a good thing, since almost no-one designs in an over-voltage crowbar circuit, so failure of a series regulator is often accompanied by wholesale destruction of the circuitry that uses the regulated supply. This is especially so with logic circuitry ... 5V logic circuits will typically suffer irreparable damage with a supply voltage above 7V.
In the circuit shown above, a simple zener is boosted (or enhanced) by adding R3 and Q1. As a quick test, the circuit was simulated. The 24V DC input was deliberately 'polluted' with a 2V peak (1.414V RMS) 100Hz sinewave to measure the ripple rejection. The circuit as shown was able to reduce the ripple from 1.4V RMS to 2mV RMS, a reduction of 56dB.
If R1 and R2 are replaced with a single 100Ω resistor (retaining C2), ripple rejection falls to 40dB (14mV RMS ripple). This technique for ripple reduction used to be very common when people built discrete regulated power supplies. The two resistors and the 470μF capacitor (C2) form a low pass filter, with a -3dB frequency of 14.4Hz. The enhanced zener performs far better than the zener diode by itself, because it introduces gain, and minimises the current through the zener diode. The bulk of the dissipated power is in Q1. Without the transistor, performance is much worse (6mV RMS ripple, 47dB attenuation at 100Hz).
The capacitor in parallel with the zener (C2) is far less effective than C1. Why? Because the zener has a low impedance (especially the enhanced version shown), this acts in parallel with the cap's impedance. Even a 470μF cap for C2 has little effect in this circuit. With no capacitors at all, the output ripple is 15mV, so C2 only reduces the ripple by less than a few microvolts. It's there to bypass the output at high frequencies.
By splitting the resistance to C1, the capacitor works with the effective impedance of the two resistors in parallel - this is much greater than the impedance of the zener, so the cap has more effect. Needless to say, a larger capacitance gives better ripple performance - doubling the capacitance halves the ripple voltage. The circuit was supplying a load current of about 60mA (12V, 200Ω load).
At full load (~60mA), the zener dissipation is under 20mW, and Q1 dissipates 270mW. This rises to over 1W with no load. If only a 1W zener were used, it would fail if the circuit were operated with no load for more than a few seconds. Resistor dissipation remains the same whether the circuit is loaded or not, but it increases if the output is shorted to ground. The two resistors need to be at least 1W, since each dissipates about 680mW.
For more information on the use of zener diodes in general, see AN008 - How to Use Zener Diodes on the ESP website. The design of shunt regulators in general isn't difficult, but there are quite a few things that need to be calculated. The unregulated input voltage must be higher than the desired output, and this includes any ripple. For example, if the minimum voltage is 13V and the maximum 17V (4V peak-to-peak of ripple) you can't expect to get 12V output because 1V headroom just isn't enough. The minimum voltage should be not less than 50% greater than the desired output. For 12V out, that means no less than 18V input, but performance will be poor with less than a 100% margin (24V in for 12V out). Remember too that the incoming mains will vary and this has to be taken into account as well.
The feed resistance (R1 and R2 in Figure 5.1) should pass a minimum of ~1.2 times the maximum load current. If your circuit draws 50mA then the resistors need to pass at least 60mA. The voltage across the feed resistance is the input voltage minus the output voltage. You then need to work out the power dissipation of the resistors, zener and shunt transistor.
Where a physically small power supply is required for a project (including audio, but not necessarily for true hi-fi use), one can use the intestines of a miniature 'plug-pack' (aka 'wall-wart') SMPS. Although only small, some of these are capable of considerable power, but installation is not for the faint-hearted. Quite obviously, the circuit board must be extremely well insulated from chassis and protected against accidental contact when the case is open.
The advantage is that the project does not require an external supply. This is often a real pain to implement, because there is always the possibility that the wrong voltage or polarity can be applied if the external supplies are mixed up (which is not at all uncommon). The disadvantage is that the unit now must have a fixed mains lead or an approved mains receptacle so a lead can be plugged in. Somewhat surprisingly, there's no requirement for 'special' approvals (as apply to all plug-pack supplies sold in Australia). Because the supply is not external, it isn't possible for anyone to come into contact with any part of it, but it will still be safe if installed into an earthed (grounded) chassis. This means a 3-pin plug - no exceptions!
That doesn't mean that you can buy any old rubbish from China - it must be a safe design, with proper insulation, filtering and all necessary EMI (electromagnetic interference) prevention measure in place. There are many supplies that are fit for one location only - the local rubbish tip! (Or preferably an electronics recycling facility.)
WARNING : The following description is for circuitry, some of which is not isolated from the mains. Extreme care is required when dismantling any external power supply, and even greater care is needed to ensure that the final installation will be safe under all foreseeable circumstances (however unlikely they may seem). All primary circuitry operates at the full mains potential, and must be insulated accordingly. It is highly recommended that the negative connection of the output is earthed to chassis and via the mains safety earth. Do not work on the power supply while power is applied, as death or serious injury may result.
The photo in Fig 5.1 shows a typical 12V 1A plug-pack SMPS board. As removed from the original housing, it has no useful mounting points, so it is necessary to fabricate insulated brackets or a sub-PCB (made to withstand the full mains voltage) to hold the PCB in position. Any brackets or sub-boards must be constructed in such a manner that the PCB cannot become loose inside the chassis, even if screws are loose or missing. Any such board or bracket must also allow sufficient creepage and clearance distances to guarantee that the primary-secondary insulation barrier cannot be breached. I shall leave the details to the builder, since there are too many possible variations to consider here.
This arrangement has some important advantages for many projects. These supplies are relatively inexpensive, and the newer ones satisfy all criteria for minimum energy consumption. Most will operate at less than 0.5W with no load, and they have relatively high efficiency (typically greater than 80% at full load). The output is already regulated, so you save the cost of a transformer, bridge rectifier, filter capacitor and regulator IC. Note that this supply used UK mains pins, and does not have Australian approval. However, it is compliant with CE regulations, it would almost certainly pass tests to AS/NZS¹ and is safe and well designed. In particular, the isolation barrier between mains and output sides is generous, and is a minimum of 6mm.
¹ AS/NZS - Australian/ New Zealand Standards
Overall, this is a far better supply than most of those available from eBay or the like, and it's small - the outside dimensions of the ½ case seen below are 65 × 39mm (the 'ears' required by UK regulations were removed). If you keep the top cover, that can be clipped back on after installation. However, getting it off again if required may pose a real challenge.
The SMPS pictured is a 12V 1A (12W) unit, and for most applications this will provide more than enough current. Consider the safety advantage compared to a transformerless supply - the finished project can have accessible inputs and outputs, and is (at least to the current standards) considered safe in all respects. Personally, I would only consider it to be completely safe if the chassis is earthed. However, it is legally allowed to be sold in Australia, and we have reasonable safety standards for external power supplies. They are 'prescribed items' under the Australian safety standards, meaning that they must be approved before they can be sold.
In some cases, the original plug-pack case may be able to be re-used. Of course, this means that you need to be careful when it's split apart, but it is possible as seen above. The two mains pins and plastic 'earth' pin were removed, and the holes for the mains pins provide convenient mounting points (check for adequate clearance, and add insulation!). The case shown has 8mm clearance below the bottom of the PCB. However, there are components under the board, so insulation is an absolute requirement. You could use plastic screws, but they aren't very strong. There are many options for mounting, so you can decide what works for you.
In the above, you can see 3mm threaded brass inserts (available from eBay for about AU$10.00 for 100 pcs.), melted into the pin holes. Because there's not a lot of plastic in this region, reinforcement with epoxy or UV (ultraviolet) cured adhesive is essential so the inserts can't be pulled out. Make sure that you don't get any glue in the threaded hole, or the insert will be ruined. The photo also shows the insulating sheet that goes under the PCB. While this is specific to the PSU I used, a similar approach can be used with any SMPS case. When building any project you need to be a bit adventurous (or inventive) to come up with a solution that's easy to put together, while still retaining the maximum safety of the end result.
There is no more effort required to install a supply such as this instead of a linear supply, and in reality there's less if you can retain (and modify) the original case. When wired up, you can safely work on the secondary side (as with a linear supply). While it might be a little more expensive than a linear supply, it's also much smaller. If you are a canny shopper, you should be able to get a supply of the type shown for about AU$10 (I got mine from Element14 for less than AU$10 at the time). It came with a UK plug, but that was irrelevant as it was never going to be plugged in.
Another possibility is a stand-alone AC/DC converter such as many advertised on eBay. The type shown doesn't come with a case, so you'll need to fabricate something, using metal (with appropriate insulation), glued plastic or 3-D printed. The boards for a 12V, 500mA versions typically measure around 52 × 24mm. These are available from China, at a cost of around AU$7.00 each including postage. Compared to the Fig. 6.1/2 versions there's a lot more messing around, as there is no case that can be re-purposed. This is still a worthwhile option though.
¹ The photo is from an eBay supplier page, and is shown for reference. It's almost impossible to describe these adequately, hence the photo. A link is pointless, because they change regularly.
In particular, look for input common-mode chokes (the dual-winding part at the top left) and an output choke (the cylindrical part between the output caps on the right). Proper filtering is essential, or the noise level will be much higher than it should be. You can't test for electrical safety unless you have access to a Megger (high voltage insulation tester), which will have an output voltage of 500V or 1kV (DC). The measured resistance between input and output should normally be at least 1,000MΩ (1GΩ), and anything less is an indication of leakage between primary and secondary. You may need to remove the Class-Y cap if it's rated as Y2 - the test voltage should be no more than 500V. No 'no-name' SMPS should ever be used unless you can verify that the insulation is sufficiently robust.
I generally test at 1kV, but keep the test duration to about 10 seconds so parts aren't stressed too much. Most supplies I have tested show 2,000MΩ (2GΩ - the upper readable limit for my tester). Interestingly (or not, depending on your perspective), one supply I measured gave a rather poor 50MΩ from input to output. This was traced to the inexplicable addition of 5 × 10MΩ SMD resistors in series, bridging the isolation barrier. Needless to say these (and their PCB pads & traces) were removed. I have no idea of why anyone thought that was a good idea.
A Megger See Note (high voltage insulation tester) is a very worthwhile piece of test gear for any hobbyist. It lets you verify that your latest creation is electrically safe (at least within the limits of the tester), and you can be fairly sure that if the insulation tester tells you that the insulation resistance is over 200MΩ at 500V DC (or 1kV DC for the paranoid), there is little likelihood of insulation failure and you haven't made any silly mistakes that could cost you your life. Most have an upper limit of 2GΩ, with some extending to over 5GΩ. An insulation tester is not a panacea though, so you must always use best wiring principles when working with mains voltages. 'Generic' high voltage testers can be obtained for around AU$60.00+ - not an especially cheap item, but if it saves your life it's a bargain!
Note: Megger® is a registered trade mark for insulation testers, but like Variac® the name has become part of the lexicon of electronics because they've been with us for so long. See Megger for the original.
Insulation tests are performed using DC. There is always some capacitance between the primary and secondary of a transformer (50/ 60Hz or SMPS), and with any SMPS there's also the Class-Y capacitor. These will give an impedance proportional to the capacitance and frequency. At 50Hz, you'd normally expect an impedance (not resistance) of around 1MΩ or more. Using DC (at a voltage ≥ the peak of the AC voltage) eliminates problems due to capacitance. Subjecting insulation to a 'hipot' (high potential) test (especially AC) is often considered destructive, and if so, the item tested must not be used after testing! The insulation may not have failed, but it has been subjected to a test voltage well beyond its design ratings, which may weaken the insulation materials.
As always, obtaining the test procedures for where you live involves getting a copy of the relevant standards documents. These are only available from the bodies that set the standards, and they are very costly. I've complained about this in several pages on the ESP site, and the situation is made worse because not everything is in one document, so you may have to purchase several standards to get all the information you need. Typically, standards documents refer to other standards documents, and you need them all to know just what is required.
Protection against accidental contact with live parts is always advised, even when the device is obviously mains powered. With any supply from China, always verify that the Y-cap (next to the transformer in the photo) really is a Class-Y component. I've seen too many supplies using 1kV ceramic caps as a substitute. If there is any doubt, replace it with a genuine Class-Y1 (or Class-Y2) certified safety capacitor.
The regulations worldwide are different, but in most cases, it's expected that one will have to use a 'tool' to gain access to live parts. A screwdriver generally counts, but as many will be aware, some manufacturers take this to extremes, using 'security' screws that require a particular tool that fits the recess. These range from Torx to more 'advanced' tools, but nothing will keep people out if they are determined enough. Many commercial SMPS use a glued case that can be difficult to get apart without damaging it beyond repair, while others use (very secure!) clips that can be undone if you know where they are and have the right tools.
One thing that you need to be aware of is that almost all modern SMPS are designed to comply with energy efficiency standards. That means that at low (or no) load, they operate in a mode commonly referred to as 'skip-cycle'. The supply will switch off for much of the time, only turning on when the output voltage falls below the threshold by a few millivolts. These have a no-load rating of less than 500mW (sometimes as low as 100mW), with the idea that they don't draw significant mains power when plugged in but unused. The regulators (world-wide) determined in their infinite wisdom (note careful use of sarcasm) that everyone leaves their power supplies plugged in, even when they aren't being used. Some people do, but many don't!
The result is that the supplies are very noisy within the audio frequency range with minimal load. I have captured the noise I found with the example supply shown above, with no load, 12mA output, 24mA output and 63mA output. If your application is audio (or within the audio frequency range, the supply is unusable unless the minimum load is drawn (typically around 100mA, but it varies). This also applies to USB chargers, so if you try to use one of those to power a project, you must ensure that you draw enough current to force the supply to operate with a 'normal' duty cycle. The clue will be that your project is noise-free when powered from the USB port on a PC, but unusable with a separate USB charger.
No Load, 242Hz
12mA Load, 1,917Hz
24mA Load, 3,267Hz
63mA Load, 4,560Hz
These noises were recorded from the SMPS output via a 15kHz low pass filter to minimise high-frequency 'hash' that would alter the recording. Each was amplified by 100 (40dB) post recording. The files are MP3 because there is no expectation of fidelity - this is stuff you don't want to hear. Unless you draw enough current (or add a very serious filter stage) you will get this noise through an audio circuit. Maybe not all, or maybe it will be amplified to make it even worse. For the frequency, only the fundamental is shown. The waveforms are roughly triangular, and contain both even and odd harmonics. With 100mA current drain, the noise was well outside the audio range (minimal or no skip-cycle behaviour). The supply you use will be different!
The sound files are each ~10s long, and have been boosted so they are louder than the direct output from the supply. I used the 12V supply described above, and captured the noise using a PC sound card. This is a very real problem, and there doesn't appear to be any form of filtering that prevents the noise from getting through. It could (probably) be done with a so-called 'capacitance multiplier' but at the expense of some voltage loss. In most cases, a resistor that draws enough current to force the SMPS into continuous operation will be the easiest - albeit wasteful of energy. At least 500mW will be needed in most cases, but up to 1W may be required.
The main purpose of this article is to provide some ways you can create a small power supply to power ancillary circuitry within a chassis. It's not a substitute for the main article that covers a much wider range and includes transformerless power supplies (see Small, Low Current Power Supplies - Part 1.
There is no doubt that the traditional transformer based supply is the safest and has the highest reliability. It is extremely easy to ensure that no live connections are accessible, often needing nothing more than some heatshrink tubing to insulate joined wires. Note that if possible, two layers of heatshrink should be used to provide reinforced insulation over joined wiring. I have linear supplies that are over 50 years old, and they remain functional to this day. The same cannot be expected of switchmode supplies! Good ones can still survive for a reasonable time, especially if they are operated in free air (without the original enclosure). The lower the operating temperature, the longer they will survive. Protection from accidental contact is very important though, and is harder with a SMPS than a simple transformer based linear supply.
A 50/ 60Hz transformer has full galvanic isolation and requires little or no EMI filtering, leakage current is extremely low, and a well made transformer based supply is so reliable that it will almost certainly outlive any equipment into which it is installed. While it's usually not the cheapest option, a transformer provides a reasonable attenuation of common mode mains noise, and the final supply can be made to be extremely quiet, with virtually no hum or noise whatsoever. No-load efficiency is not as good as a modern SMPS, but the 'wasted' power is generally no more than a couple of watts. Yes, you pay for it, but it won't be noticed on your electricity bill.
The next best option is a modified plug-pack SMPS or a purpose built chassis mounting SMPS. These are useful where high efficiency is needed, along with very low standby power requirements. They are rather (electrically) noisy though, and the full range of voltages is not available. Where possible, design circuits to suit available voltages (12V is always a safe bet, and that's used throughout this article), rather than trying to find a supply that provides an 'odd' voltage. An example is 30V - it's a nice round number, but try to get a 30V supply that you don't have to build yourself!
| Main Index
Articles Index
|
