Generating +/- 14V from 3V supply

[new299]

I need to generate +/- 15V to supply a DAC and associated opamps. I have a 3v supply voltage, and ideally I lead a low noise solution (but am not concerned about efficiency). It needs to supply about 100mA.

What should I use? Art of Electronics seems to suggest a low-power step-up converter. So a boost converter? Can anybody recommend a suitable low noise part?

Should I follow the boost converter with a linear regulator? So perhaps boost to 20v and then use a 15v regulator to give a low noise output?

Are there other low noise configurations available to me that I might have missed?

[retrolefty]

A bigger problem is generating the negative voltage as most switcher modules are not isolated so as to be able to output negative voltage when supplied with a positive input voltage. There are special isolated 'power modules' that will generate both + and - rail voltage isolated from the input voltage. They are not cheap at the 100 ma and higher current values.

[new299]

A bigger problem is generating the negative voltage as most switcher modules are not isolated so as to be able to output negative voltage when supplied with a positive input voltage. There are special isolated 'power modules' that will generate both + and - rail voltage isolated from the input voltage. They are not cheap at the 100 ma and higher current values.

Thanks, do you have a link to such a module?

To generate a negative voltage can I use an inverting buck-boost topology? (http://www.ti.com/lit/an/slyt286/slyt286.pdf)

[retrolefty]

A bigger problem is generating the negative voltage as most switcher modules are not isolated so as to be able to output negative voltage when supplied with a positive input voltage. There are special isolated 'power modules' that will generate both + and - rail voltage isolated from the input voltage. They are not cheap at the 100 ma and higher current values.

Thanks, do you have a link to such a module?

To generate a negative voltage can I use an inverting buck-boost topology? (http://www.ti.com/lit/an/slyt286/slyt286.pdf)

 On a check at one supplier I couldn't find any that work at 3vdc input, just +5vdc or higher.

 It's not what kind of switcher circuit you use, The issue is if it's negative input is isolated from it's output negative. If not you can't use two of them to generate two bipolar output voltages.

[new299]

A bigger problem is generating the negative voltage as most switcher modules are not isolated so as to be able to output negative voltage when supplied with a positive input voltage. There are special isolated 'power modules' that will generate both + and - rail voltage isolated from the input voltage. They are not cheap at the 100 ma and higher current values.

Thanks, do you have a link to such a module?

To generate a negative voltage can I use an inverting buck-boost topology? (http://www.ti.com/lit/an/slyt286/slyt286.pdf)

 On a check at one supplier I couldn't find any that work at 3vdc input, just +5vdc or higher.

 It's not what kind of switcher circuit you use, The issue is if it's negative input is isolated from it's output negative. If not you can't use two of them to generate two bipolar output voltages.

Looks like this might be an option? NTA0315MC (http://www.digikey.co.uk/scripts/DkSearch/dksus.dll?Detail&itemSeq=177564871)

The inverting buck-boost topology shown in AoE appears to suggest that it inverts the positive input with respect to system ground. Have I misunderstood?

[edavid]

If you want to put together modules, it's probably easiest to use a 3V to 5V stepup followed by a 5V to +/- 15V.

If you want to build it yourself, you could use a 3V powered sine wave oscillator that drives a stepup transformer and linear postregulators... low noise, inefficient, easy to build.

If you could cut down the output power a bit, you would have an easier time.

[new299]

If you want to put together modules, it's probably easiest to use a 3V to 5V stepup followed by a 5V to +/- 15V.

If you want to build it yourself, you could use a 3V powered sine wave oscillator that drives a stepup transformer and linear postregulators... low noise, inefficient, easy to build.

If you could cut down the output power a bit, you would have an easier time.

Thanks, this sounds like an interesting option. Would you suggest an opamp based sine wave oscillator?

If I could reduce the output power, what other options would be available to me?

[Bud]

Check LM3224, it is specked down to 2.7V. The datasheet shows how to generate a negative rail.

I would not chain Dc-Dc converters, i tried, too much noise. The downstream converter adds to its output noise from upstream converter.

[langwadt]

A bigger problem is generating the negative voltage as most switcher modules are not isolated so as to be able to output negative voltage when supplied with a positive input voltage. There are special isolated 'power modules' that will generate both + and - rail voltage isolated from the input voltage. They are not cheap at the 100 ma and higher current values.

Thanks, do you have a link to such a module?

To generate a negative voltage can I use an inverting buck-boost topology? (http://www.ti.com/lit/an/slyt286/slyt286.pdf)

 On a check at one supplier I couldn't find any that work at 3vdc input, just +5vdc or higher.

 It's not what kind of switcher circuit you use, The issue is if it's negative input is isolated from it's output negative. If not you can't use two of them to generate two bipolar output voltages.

you can "cheat" and get both positive and negative from a standard boost converter

http://www.maximintegrated.com/en/images/appnotes/4497/4497Fig01.gif

you can also use a dual inductor, http://www.maximintegrated.com/en/images/appnotes/660/660Fig02.gif

if you add linear regulators it doesn't a matter so much the negative voltage isn't tightly regulated

[MagicSmoker]

You have a 3V supply - that means your converter needs to use bipolar junction transistors (BJTs).
You want low noise - that suggests a converter topology that features sinusoidal currents and/or voltages

The simplest circuit that meets both of the above conditions is a current-fed Royer. This is a push-pull converter with a choke input and a resonating capacitor in parallel with the primary. Not really a "Beginners" type of thing to design mainly because of the transformer but there aren't too many other choices that will work at 3V.

[Jeroen3]

How low noise?
You can get those isolated unregulated dc/dc's from 3.3 to 24Volts. Add some of your own regulators with nice filtering, and you're good to go, right? Since they are isolated you can tie two together to get +-24V. Instead of buying the +-24V version.

[new299]

Check LM3224, it is specked down to 2.7V. The datasheet shows how to generate a negative rail.

I would not chain Dc-Dc converters, i tried, too much noise. The downstream converter adds to its output noise from upstream converter.

Thanks, I found the description of negative rail generation in the datasheet a bit confusing, but I will re-read it. However it seems to note that the negative rail is unregulated? If you know of a simpler example circuit than that shown in the datasheet that would help me I think.

[new299]

I would not chain Dc-Dc converters, i tried, too much noise. The downstream converter adds to its output noise from upstream converter.

Does this mean you wouldn't recommend a switching converter followed by a linear regulator. If I did this the linear regulator wouldn't actually help?

[MagicSmoker]

I would not chain Dc-Dc converters, i tried, too much noise. The downstream converter adds to its output noise from upstream converter.

Does this mean you wouldn't recommend a switching converter followed by a linear regulator. If I did this the linear regulator wouldn't actually help?

No, Bud meant that cascading switching converters results in more noise; cascading a switching converter with a linear regulator will result in lower noise.

But as I implied in my post above, you aren't going to find too many (if any!) COTS* dc-dc converters that will accept a nominal 3V input (presumably from 2 1.5V alkaline cells) and deliver +/-15V at 3W total output, regardless of their noise spec.

Now all that said, I actually question your requirements here because many op-amps have excellent PSRR** well past the audio range and so don't really need low-noise supplies. The DAC may be a different story, but usually some simple LC filtering of the supply rails will suffice as long as the reference is very quiet and accurate.


* - Commercial Off The Shelf
** - Power Supply Rejection Ratio

[ez24]

Here is an interesting chip if you can use +- 12 or +- 15 

http://www.aliexpress.com/item/DC-DC-Isolated-Power-Module-24V-dc-buck-converter-turn-negative-15V6W-Manufacturers-selling-special/1746272285.html

Look at the data sheet, not the chip.  There is one that can go from 4.5v to  +- 12  or  +-15v
Of course you have to find a way to go from 3 to 4.5v.

Here is a dual linear module (you may need this if you have to have +-14v):

http://www.aliexpress.com/item/Cheapest-Top-qty-LM317T-LM337-positive-and-negative-dual-power-adjustable-power-supply-board/32311088563.html

But the ad is confusing and says it cannot step up.  So you may need 16 vdc ??

Art of Electronics says you need a linear reg on the end of a switcher to reduce noise


Curious on what you come up with.

[edy]

I'm curious as to whether this scheme will work although I'm sure it is not an efficient design. I think it was suggested earlier.

Take a 3V DC source, converted to AC and drive a transformer with a central tap that becomes the new ground, and ends become + and - rails, which need to be then rectified. I'm not exactly sure how the central tap needs to be rectified or whether it always stays at zero compared to the 2 ends or whether it swings from + to - also at the AC frequency.

Another question is can you make the AC by driving a 555 timer to produce a 50-50 square wave? Once it gets to the transformer and you tap the center it will have theoretically shifted the "zero volt" reference to the tap anyways, even though incoming signal on primary side is ranging from 0 v to some positive voltage?

[Jeroen3]

I've used an sn6501 for that exact purpose several times. Why even bother thinking of 555.
However, not for 100's of mA.
It's a great method of creating +-5v, or isolated 5v.

[krivx]

100mA is just about the top end for switched capacitor converters (eg http://www.maximintegrated.com/en/products/power/charge-pumps/MAX1044.html) which can invert to provide a bipolar supply. If you can manage a 3->15V boost converter this might work to provide the negative supply.

[Bud]

Check LM3224, it is specked down to 2.7V. The datasheet shows how to generate a negative rail.

I would not chain Dc-Dc converters, i tried, too much noise. The downstream converter adds to its output noise from upstream converter.

Thanks, I found the description of negative rail generation in the datasheet a bit confusing, but I will re-read it. However it seems to note that the negative rail is unregulated? If you know of a simpler example circuit than that shown in the datasheet that would help me I think.

The negative rail is derivied with just a plain vanilla halfwave voltage doubler using two diodes and two capacitors  (C1,C2,D2,D3). The negative voltage will be smaller than the positive one because of the voltage drop across the two diodes, and drop more as you draw more current, so one would need  to have a linear LDO regulator to stabilize it, providing the doubler can handle the load current. If you need 100mA on the negative rail that topology would not probably work. In one of my designs I just draw about 10mA from it (via a cascaded LDO) and it has no problems whatsoever.

[new299]

100mA is just about the top end for switched capacitor converters (eg http://www.maximintegrated.com/en/products/power/charge-pumps/MAX1044.html) which can invert to provide a bipolar supply. If you can manage a 3->15V boost converter this might work to provide the negative supply.

I put together the attached design as a first pass, using an LM3224 to go to 16v, then a ltc1144 to invert that. This then goes through TPS7A49 and TPS7A30 regulators, which are billed as "low noise".

Does this seem reasonable? Do the parts selected seem like reasonable choices?

[ez24]

The title of this post says +- 14v but your schematic shows +- 15v

If really +-15v I suggest you change the title and you may get more hits

[new299]

Your LT chip is only good to about 50mA depending on the quality of load regulation you are after, it has about 56 oHm output R so it drops 2.8V under 50mA load at 25C. You can expect much worse with increasing temperature.

...

I've done that with bucks but apparently it doesn't work as well with boost without custom magnetics. Here is an application note from TI with a capacitive coupling for aux negative voltage with reasonable regulation to 1.5W.

http://www.ti.com/lit/an/slua288/slua288.pdf

You can also do coupled inductors but for it to work well you would have to wind your own.

Thanks for pointing that out. It looks like the MAX889 might be suitable?

http://datasheets.maximintegrated.com/en/ds/MAX889.pdf

[new299]

I worked up this design and wrote up my final notes here: http://41j.com/blog/2015/08/15v-and-5v-ref-board/. Fingers crossed.

[ez24]

I do not get it.  The MAX889 is spec'd to -6 - how do you get -15v ?

[new299]

I do not get it.  The MAX889 is spec'd to -6 - how do you get -15v ?

You're right, I screwed up. I'll have to try something else...

[ez24]

ha ha
The first time I have been right on this forum 

[new299]

ha ha
The first time I have been right on this forum 

Ah well, at least I made somebody happy. Looks like the LT1054 might be suitable...

[flynwill]

I recently built such a thing with the LT1372, It probably falls short of 100 ma with 3V input (It'll easily get there with 4.5-5 V input).  I just used an additional LC trap to further suppress the switching ripple.

[new299]

As an update on this I build a board around the LM3224 and LT1054 and it appears to be working well. Wrote up some notes here:

http://41j.com/blog/2015/08/13v-psu-board-bring-up/

[Beninot]

Hi, I don't know if this may help you, but the way I found to generate +/- voltage with simple variables voltage regulators like the lm317, is to get a lm 317 to generate a voltage that is half the voltage of your supply and then, you can use it as a "virtual ground", the voltage between it and your positive will be +1/2 your input and for the negative, it will be -1/2 your input

Hope it can help you!   

[new299]

Hi, I don't know if this may help you, but the way I found to generate +/- voltage with simple variables voltage regulators like the lm317, is to get a lm 317 to generate a voltage that is half the voltage of your supply and then, you can use it as a "virtual ground", the voltage between it and your positive will be +1/2 your input and for the negative, it will be -1/2 your input

Hope it can help you!   

That's an interesting idea thanks! In my case I also needed to boost from 3v. I guess I could boost from 3v to 28v and then split it was the lm317. It could be an option perhaps.

[fivefish]

Can you post scope shots of your +/- DC supply? How clean are the outputs? Ripple, etc.

[edavid]

Hi, I don't know if this may help you, but the way I found to generate +/- voltage with simple variables voltage regulators like the lm317, is to get a lm 317 to generate a voltage that is half the voltage of your supply and then, you can use it as a "virtual ground", the voltage between it and your positive will be +1/2 your input and for the negative, it will be -1/2 your input

This is a bad idea, unless you can guarantee that the upper part of the circuit always passes less current than the lower part.  The LM317 can't sink current, so if the upper part of the circuit passes too much current, the "virtual ground" will be pulled up.

[Zero999]

Where are you getting the 3V from?

You want an output power of 3W so at an efficiency of 80% that's an input current of 1.25A.

[new299]

Can you post scope shots of your +/- DC supply? How clean are the outputs? Ripple, etc.

It's very noisy, so I guess I need to work on that. The transients seem to dominate, they're about 50mV! What can I do to reduce this? Any suggestions out there?

[fivefish]

Can you post scope shots of your +/- DC supply? How clean are the outputs? Ripple, etc.

It's very noisy, so I guess I need to work on that. The transients seem to dominate, they're about 50mV! What can I do to reduce this? Any suggestions out there?

Can you post it anyway. I know there'll be noise. I'm really curious and thinking of doing a similar thing.  So even with the linear regulator after the DC converter, it still has some transients? 

Have you tried a pi filter at the outputs?

[new299]

Can you post scope shots of your +/- DC supply? How clean are the outputs? Ripple, etc.

It's very noisy, so I guess I need to work on that. The transients seem to dominate, they're about 50mV! What can I do to reduce this? Any suggestions out there?

Can you post it anyway. I know there'll be noise. I'm really curious and thinking of doing a similar thing.  So even with the linear regulator after the DC converter, it still has some transients? 

Have you tried a pi filter at the outputs?

Attached. I'm not 100% confident in my test setup (I've read I should really shield the board?) but it seems like there's so much noise it's not worth trying that yet. It doesn't feel like the regulators do a not from probing around.

[Beninot]

Hi, I don't know if this may help you, but the way I found to generate +/- voltage with simple variables voltage regulators like the lm317, is to get a lm 317 to generate a voltage that is half the voltage of your supply and then, you can use it as a "virtual ground", the voltage between it and your positive will be +1/2 your input and for the negative, it will be -1/2 your input

This is a bad idea, unless you can guarantee that the upper part of the circuit always passes less current than the lower part.  The LM317 can't sink current, so if the upper part of the circuit passes too much current, the "virtual ground" will be pulled up.

Sorry I'm a beginer, this is the way I figured out to do this, but do you know a voltage regulator witch works at 12v and can sink current?

[edavid]

Hi, I don't know if this may help you, but the way I found to generate +/- voltage with simple variables voltage regulators like the lm317, is to get a lm 317 to generate a voltage that is half the voltage of your supply and then, you can use it as a "virtual ground", the voltage between it and your positive will be +1/2 your input and for the negative, it will be -1/2 your input

This is a bad idea, unless you can guarantee that the upper part of the circuit always passes less current than the lower part.  The LM317 can't sink current, so if the upper part of the circuit passes too much current, the "virtual ground" will be pulled up.

Sorry I'm a beginer, this is the way I figured out to do this, but do you know a voltage regulator witch works at 12v and can sink current?

TLE2426 is good for low power circuits, otherwise it makes more sense to use a conventional split supply.

[new299]

Well, my current guess is that the transients are due to poor layout? (current layout attached) In particular that the feedback resistors are very close to the inductor (as shown in the attached). I kind of assume EMF from the inductor effects the feedback network and results in the spikes I'm seeing?

So I plan to revise the layout and see if that helps. But any suggestions would be very welcome.

[MagicSmoker]

It's very noisy, so I guess I need to work on that. The transients seem to dominate, they're about 50mV! What can I do to reduce this? Any suggestions out there?

Ahem...

You have a 3V supply - that means your converter needs to use bipolar junction transistors (BJTs).
You want low noise - that suggests a converter topology that features sinusoidal currents and/or voltages

The simplest circuit that meets both of the above conditions is a current-fed Royer. This is a push-pull converter with a choke input and a resonating capacitor in parallel with the primary. Not really a "Beginners" type of thing to design mainly because of the transformer but there aren't too many other choices that will work at 3V.

[fivefish]

Magic, are you referring to the type of DC-DC converters that require the use of small transformers with primary and secondary windings?  The secondary is then DC rectified and filtered?

[SaabFAN]

Well, my current guess is that the transients are due to poor layout? (current layout attached) In particular that the feedback resistors are very close to the inductor (as shown in the attached). I kind of assume EMF from the inductor effects the feedback network and results in the spikes I'm seeing?

So I plan to revise the layout and see if that helps. But any suggestions would be very welcome.

Have you tried to add a LC-Filter to the output?

[RoadRunner]

i suggest LT3463 - Dual Micropower DC/DC Converters do very good job. i have used them for e-ink bias circuit.

http://www.linear.com/product/LT3463

[rs20]

I would not chain Dc-Dc converters, i tried, too much noise. The downstream converter adds to its output noise from upstream converter.

Citation needed? Sorry to dig up this old post, but how can this be? Put simplistically, the switching noise from the first regulator has to make it through through at least two stages of smoothing, so even if the two sources of noise are added, I would think the first source of noise is heavily attenuated. I grant there could be subtleties, like perhaps the control loop of the second converter being upset by noise from the first... but at the end of the day, the second converter is slamming the output of the second inductor from 5V down to 0V up and down again continually, how can adding a ~0.1V of ripple to that 5V be of any relevance?

[fivefish]

Citation needed? Sorry to dig up this old post, but how can this be? Put simplistically, the switching noise from the first regulator has to make it through through at least two stages of smoothing, so even if the two sources of noise are added, I would think the first source of noise is heavily attenuated.

I've seen it.

Example: 12VDC switching wall wart (typical china stuff you can get from eBay), powering a 12V to <whatever> DC-DC switching converter circuit.  Scope the output of the 2nd converter and you'll see switching spikes caused by/originating from the 12VDC wall wart.  Unplug the wall wart, the switching spikes disappear, and for a few seconds you have "cleaner" power until the wall wart capacitor charge runs out.

[rs20]

In particular that the feedback resistors are very close to the inductor (as shown in the attached). I kind of assume EMF from the inductor effects the feedback network and results in the spikes I'm seeing?

The spikes you are seeing are happening at over 1 MHz -- so the spikes you are seeing are surely just switching noise? There's no way that the feedback resistor and DC-DC controller are responding so fast that they are creating the spikes you see. In any case, the little spikes you see are less than 20 nanoseconds long, so simple passive filtering as suggested by SaabFAN is surely the cure you need. But, why you'd want to carefully smooth out the voltage input to a switching converter is beyond me. Save all that trouble for after the converters, even if the spikes make it through the converters, they can't be any worse. Might as well deal with the noise from your supply, and the noise from your switching supplies, all at once -- with passive filtering and maybe linear regulation.

Citation needed? Sorry to dig up this old post, but how can this be? Put simplistically, the switching noise from the first regulator has to make it through through at least two stages of smoothing, so even if the two sources of noise are added, I would think the first source of noise is heavily attenuated.

I've seen it.

Example: 12VDC switching wall wart (typical china stuff you can get from eBay), powering a 12V to <whatever> DC-DC switching converter circuit.  Scope the output of the 2nd converter and you'll see switching spikes caused by/originating from the 12VDC wall wart.  Unplug the wall wart, the switching spikes disappear, and for a few seconds you have "cleaner" power until the wall wart capacitor charge runs out.

How did the amplitudes of the two sources of noise compare? That's weird. That implies the duty cycle of the second converter isn't paying much attention to the output. What was the topology of the second converter? It's quite clear that a Constant-on-time controller cannot exhibit this behaviour, nor a hysteretic controller.

The bottom line is that the second DC-DC converter hacks up whatever input voltage it gets into a massively rippley square wave. How input spikes survive the smoothing that removes all the squareness is hard to comprehend. If you're seeing poor layout allowing coupling from the input directly to the output, or common-mode noise, those are different issues entirely.

[new299]

It's very noisy, so I guess I need to work on that. The transients seem to dominate, they're about 50mV! What can I do to reduce this? Any suggestions out there?

Ahem...

You have a 3V supply - that means your converter needs to use bipolar junction transistors (BJTs).
You want low noise - that suggests a converter topology that features sinusoidal currents and/or voltages

The simplest circuit that meets both of the above conditions is a current-fed Royer. This is a push-pull converter with a choke input and a resonating capacitor in parallel with the primary. Not really a "Beginners" type of thing to design mainly because of the transformer but there aren't too many other choices that will work at 3V.

Thanks, I'd like to see how far I can push the boost based design at the moment. It's helping me understand the limitations of boost converters anyway I think.

[new299]

The spikes you are seeing are happening at over 1 MHz -- so the spikes you are seeing are surely just switching noise? There's no way that the feedback resistor and DC-DC controller are responding so fast that they are creating the spikes you see. In any case, the little spikes you see are less than 20 nanoseconds long, so simple passive filtering as suggested by SaabFAN is surely the cure you need. But, why you'd want to carefully smooth out the voltage input to a switching converter is beyond me. Save all that trouble for after the converters, even if the spikes make it through the converters, they can't be any worse. Might as well deal with the noise from your supply, and the noise from your switching supplies, all at once -- with passive filtering and maybe linear regulation.

I'll tried to add LC filtering and it's not seemed to help. I'll try fiddling some more.

One thing I'm confused about... If I disconnect the boost converter output from the linear regulator at it's output, I still see the transient noise on the regulator output. Is this then likely EMI, noise on the supply (I tried adding LC filtering here too though), or is it a measurement artifact?

[MagicSmoker]

Magic, are you referring to the type of DC-DC converters that require the use of small transformers with primary and secondary windings?  The secondary is then DC rectified and filtered?

Yes. More specifically, the original Royer is a self-oscillating voltage-fed push-pull converter that delivers a nearly-pure DC output voltage even before any filtering. The biggest downside of the original Royer is its low efficiency (due to the high peak current in the transistors when the transformer saturates each half cycle), but this can be fixed by simply inserting an inductor in series with the primary center-tap (and flyback diodes across each BJT), which turns it into a current-fed self-oscillating push-pull converter. Then for the ultimate in noise reduction you can place a resonating capacitor across the primary which changes the voltage waveform from a square wave into a sine wave.

The resonant current-fed push-pull was very popular for CCFL backlights in laptops (which have since been rendered obsolete by white LEDs) so a lot of literature will be aimed at that application, but the principles are the same whether the output is in volts or kilovolts.

[MagicSmoker]

...
Thanks, I'd like to see how far I can push the boost based design at the moment. It's helping me understand the limitations of boost converters anyway I think.

Okay, then to minimize switching noise you'll want to use a large value of inductance (to ensure the converter operates in continuous conduction mode [aka - CCM]), a Schottky output diode (almost mandatory for CCM) and two stages of filtering comprised first of a small value, good quality ceramic capacitor (like X5R or X7R dielectric) in parallel with the usual large value electrolytic followed by a lossy ferrite bead* and another good quality ceramic capacitor to integrate away any spikes.

* - acceptable at low current; not a trick that scales up well, however.

[edit - corrected parenthetical definition of CCM as "continuous conduction mode" rather than "...current..."]

[rs20]

Okay, then to minimize switching noise you'll want to use a large value of inductance (to ensure the converter operates in continuous conduction mode [aka - CCM]), a Schottky output diode (almost mandatory for CCM) and two stages of filtering comprised first of a small value, good quality ceramic capacitor (like X5R or X7R dielectric) in parallel with the usual large value electrolytic followed by a lossy ferrite bead* and another good quality ceramic capacitor to integrate away any spikes.

+1, and also make sure that you oscilloscope probe ground lead is physically connected directly to one side of the output capacitor, and the probe itself as close as possible to the other.

[new299]

Okay, then to minimize switching noise you'll want to use a large value of inductance (to ensure the converter operates in continuous conduction mode [aka - CCM]), a Schottky output diode (almost mandatory for CCM) and two stages of filtering comprised first of a small value, good quality ceramic capacitor (like X5R or X7R dielectric) in parallel with the usual large value electrolytic followed by a lossy ferrite bead* and another good quality ceramic capacitor to integrate away any spikes.

+1, and also make sure that you oscilloscope probe ground lead is physically connected directly to one side of the output capacitor, and the probe itself as close as possible to the other.

For the next revision, I've put an SMA connector on the board and go from the board to the scope directly. Would this be suitable? Or are there potential issues.

[rs20]

Okay, then to minimize switching noise you'll want to use a large value of inductance (to ensure the converter operates in continuous conduction mode [aka - CCM]), a Schottky output diode (almost mandatory for CCM) and two stages of filtering comprised first of a small value, good quality ceramic capacitor (like X5R or X7R dielectric) in parallel with the usual large value electrolytic followed by a lossy ferrite bead* and another good quality ceramic capacitor to integrate away any spikes.

+1, and also make sure that you oscilloscope probe ground lead is physically connected directly to one side of the output capacitor, and the probe itself as close as possible to the other.

For the next revision, I've put an SMA connector on the board and go from the board to the scope directly. Would this be suitable? Or are there potential issues.

I'd call that overkill -- you're ultimately trying to get this power to a whole bunch of op amps and DACs and so on, so making a circuitboard that delivers clean power to just one SMA port isn't particularly useful for your end goal. Arranging for a good shielding solution instead, if that's deemed necessary, will give an improved waveform on your scope, and better performance for your end product.

[new299]

Okay, then to minimize switching noise you'll want to use a large value of inductance (to ensure the converter operates in continuous conduction mode [aka - CCM]), a Schottky output diode (almost mandatory for CCM) and two stages of filtering comprised first of a small value, good quality ceramic capacitor (like X5R or X7R dielectric) in parallel with the usual large value electrolytic followed by a lossy ferrite bead* and another good quality ceramic capacitor to integrate away any spikes.

Thanks, from my reading of the datasheet it looks like I should use ~100uH to run in CCM, so I'll try that.

For the filtering should I also try adding a pi filter? I tried this using a 10uH inductor but it didn't seem to help much.

Does it also make sense to separate the ground plane of the switching and non-switching parts? Some of the noise I'm seeing seems to be coupled through the ground as far as I can tell (as I still see noise on the linear regulators if only the ground is connected) would coupling the switching/linear parts of the circuit via an inductor be a sane thing to do?

[new299]

Magic, are you referring to the type of DC-DC converters that require the use of small transformers with primary and secondary windings?  The secondary is then DC rectified and filtered?

Yes. More specifically, the original Royer is a self-oscillating voltage-fed push-pull converter that delivers a nearly-pure DC output voltage even before any filtering. The biggest downside of the original Royer is its low efficiency (due to the high peak current in the transistors when the transformer saturates each half cycle), but this can be fixed by simply inserting an inductor in series with the primary center-tap (and flyback diodes across each BJT), which turns it into a current-fed self-oscillating push-pull converter. Then for the ultimate in noise reduction you can place a resonating capacitor across the primary which changes the voltage waveform from a square wave into a sine wave.

The resonant current-fed push-pull was very popular for CCFL backlights in laptops (which have since been rendered obsolete by white LEDs) so a lot of literature will be aimed at that application, but the principles are the same whether the output is in volts or kilovolts.

While I wait on parts/boards for the switching design I plan to try out the Royer circuit too. I think I understand its basic operation seems fun! So now I'm trying to search out a transformer to use.

Digikey has some listed as "for CCFL power supplies". But these tend to be quite low current (30mA). Do you have any particular recommendations?

[MagicSmoker]

...
+1, and also make sure that you oscilloscope probe ground lead is physically connected directly to one side of the output capacitor, and the probe itself as close as possible to the other.

This is a really good point that I should have mentioned! The usual 10cm ground lead with alligator clip is all but useless when scoping switchmode converter circuits because it picks up radiated magnetic fields. The OP really needs to use a spring tip, like in the picture below:

[MagicSmoker]

...Thanks, from my reading of the datasheet it looks like I should use ~100uH to run in CCM, so I'll try that.

The primary factors that determine whether a switchmode converter operates in CCM or CrCM (Critical Conduction Mode, aka BCM, or Boundary Conduction Mode) or DCM (Discontinuous Conduction Mode) are the load current and switching frequency; there is no single value of inductance, then, that guarantees CCM. If a converter has to contend with a wide range of loads then it will be difficult if not impossible to ensure CCM (exception: the synchronous variants of the classical switchmode topologies).

For the filtering should I also try adding a pi filter? I tried this using a 10uH inductor but it didn't seem to help much.

No, I already gave you a super secret, hard-learned tip before which is to split your filtering up into two stages - the primary filter, which consists of the switchmode inductor and the output capacitor, followed by a secondary "spike/ringing" filter which consists of a lossy ferrite bead and a really low-loss capacitor. Using a good quality inductor for the second filter often results in worse noise.

Does it also make sense to separate the ground plane of the switching and non-switching parts?...

Separating ground planes (and joining them together at one point via a trace, a lossy ferrite bead, etc.) is kind of an advanced topic that is rarely necessary and exceptionally difficult to do properly. Usually it is sufficient to make the 2nd layer of a 4 layer board (or the bottom layer of a 2 layer board) a solid ground plane and then physically separate the noisy stuff (like a switchmode power supply) from any low-noise analog stuff.

[new299]

For the filtering should I also try adding a pi filter? I tried this using a 10uH inductor but it didn't seem to help much.

No, I already gave you a super secret, hard-learned tip before which is to split your filtering up into two stages - the primary filter, which consists of the switchmode inductor and the output capacitor, followed by a secondary "spike/ringing" filter which consists of a lossy ferrite bead and a really low-loss capacitor. Using a good quality inductor for the second filter often results in worse noise.

Ok, thank you for your help, it's extremely useful. "lossy" would mean higher resistance? 30Ohms? I don't have a good feel for that.

I kind of feel like the ferrite bead is absorbing the EMI? I guess I need to read up more of higher frequency transmission, but my current model is that significant potion of the transmission occurs in the EMF around the wire and the ferrite bead absorbs this.

[MagicSmoker]

...
Digikey has some listed as "for CCFL power supplies". But these tend to be quite low current (30mA). Do you have any particular recommendations?

You can certainly reuse the core from a COTS (commercial off-the-shelf) CCFL transformer, but all of the windings (except, maybe, the feedback) are going to have the wrong number of turns for your application and the bobbin will be divided into several sections. The nice thing about the Royer is that almost any transformer will work as long as it doesn't have too much of an air-gap (that is, does not have too-soft of a saturation knee). The frequency it will oscillate at is determined by the core area, number of turns and saturation flux density. I would buy a small ferrite toroid and some #30 wire-wrap wire (and a stripper for it, if necessary; pro-tip: most wire-wrapping tools come with a stripper) and start tinkering. Begin with about 1 turn per volt for each winding (this also applies to each half of any center-tapped windings), assume 4V of base drive for the BJTs and tweak as necessary. So, for example, you might have a primary that is 3T x 2 (ie - 6CT - for 6 turns center tapped) a 4T feedback winding, and a 30CT secondary.

..."lossy" would mean higher resistance? 30Ohms? I don't have a good feel for that.

"Lossy" refers to the ferrite core material (and the resistance of the windings) which results in an inductor with low Q, which is good for filtering; bad for power conversion or tuned circuits.

One other thing with regards to the boost converter - the inductor must be a closed-form shape. For example an E-core or toroid is good, but a rod (or air core) is bad. This is to prevent spraying magnetic fields everywhere.