EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: drwho9437 on February 03, 2023, 01:17:50 am
-
Most switching bench supplies are very noisy. Even supplies with digital monitoring can have unwanted noise. I remember in my first group we only used power supplies with panel meters.
For people who have done ultra-low noise measurements (particularly 1/f performance < 100 Hz); what PSUs have you found the best?
<350 uVrms for sure 20 MHz seem easy. How about 10 uVRMS with 1 nV/rHz down to 5 Hz or something?
What is out there?
-
Long ago when working below 10Hz we employed various type batteries for our low current , noise and frequency use.
Best
-
Are you looking at new , or old.
For new Stanford Research DC205 have low noise.
For used HP precision supplies , and certain Power Designs precision models are the way to go.
For the lowest noise...batteries , and there can still be noise.
-
https://www.youtube.com/watch?v=9JinSfCKuNQ (https://www.youtube.com/watch?v=9JinSfCKuNQ)
-
I just build my own when needed. Enclosing a regulator within the feedback loop of a precision low noise operational amplifier and using a low noise reference will easily achieve the noise of the reference alone while keeping the failsafe operation of the regulator. Lower noise then requires summing the output of multiple low noise references. Flicker noise is dominated by the reference as well, so using a chopper stabilized operational amplifier provides only marginal benefit.
Using battery power avoids noise from common mode capacitive coupling across the transformer of an off-line power supply, however some low noise off-line power supplies have good enough electrostatic shielding in their transformer and design to also be suitable.
Typical parts are any 3 terminal regulator, a precision low noise operational amplifier like the LT1001, and a low noise reference like the LM329. Buried zener reference have lower noise than bandgap references when their higher reference voltage is considered.
-
The few times I've needed very low noise, I've just used batteries. I was working for Harris in a new group that hadn't gotten much equipment yet, and I found that noise from my switching DC supply (all I had) was causing me fits in an RF synthesizer and modulator I was working on. I went to the dollar store at lunch time and found something that had batteries in it (a light saber that had three stacked button cells), and used that to get through the moment. I still have the light saber to tell the story.
Reading David Hess's reply makes me want to cobble something together, but I've always just gotten through the moment with batteries. I don't normally need really low noise, and for most things a linear bench supply is fine. My linear HP6627A supplies are fine for most everything I do (quad supply, GPIB controlled, <$300 on eBay). Loud, but fantastic deal. Don't use a switched supply. I have a linear Circuit Specialists 18V/2A which I think was about $50 that is usually the first thing I use at my home bench because it doesn't make noise like my HP66xx supplies.
-
When I needed extremely noise (making measurements inside a shielded room), I used 6 V "lantern batteries" similar to this one:
https://www.bhphotovideo.com/c/product/1206290-REG/rayovac_945r_6v_heavy_duty_lantern_battery.html?ap=y&smp=y&srsltid=Ad5pg_Et_m2Lz0I9nOWeJp1Olk0JPS6ZMwhM1NAGFtfD-bqQk0OVHJHbsz4 (https://www.bhphotovideo.com/c/product/1206290-REG/rayovac_945r_6v_heavy_duty_lantern_battery.html?ap=y&smp=y&srsltid=Ad5pg_Et_m2Lz0I9nOWeJp1Olk0JPS6ZMwhM1NAGFtfD-bqQk0OVHJHbsz4)
Most batteries in this shape use spring terminals (to fit inside traffic barriers, etc.), but some have screw terminals like this one.
-
Switchers are not used for low noise app.
Just a normal well designed linear:
120/240 mains>>transformer>>LV rect/filter>>DC LDO reg>>Bypass caps
See the Linear Tech and Jim Willimas app notres and Vreg ICs for low noise DC regulators.
In metric apps a double regulator is used .
The transformer can have single or double Faraday shield.
Best,
Jon
-
120/240 mains>>transformer>>LV rect/filter>>DC LDO reg>>Bypass caps
Don't forget to add 10nf or 100nf caps in parallel, directly tied with the 4 diodes used in the rectification.
This removes the switching EMI noise of the diodes which may propagate through your system.
EG: the 1N400x series diodes has a fast enough switch speed that if you unfortunately have just the wrong trace lengths to your main filter cap, or the cap itself just has an annoyance inductance characteristic, you will see either 2 or 4 tiny spikes along the AC waveform cycle at such a frequency that these spike signals go right through your lineal regulator.
The parallel 10nf or 100nf caps right at the diodes ensures this signal is not generated regardless of the rest of your PCB / part selection design.
-
Low noise power supply topic has been discussed many times, but it would be great to show such a practical proven design with easy to get parts, indeed. People do use 317 or 723 when talking low noise, or some special ADI/TI parts. The version with 317 as the pass element and an opamp with a low noise vref - I've seen it in the DS of an opamp (308 or 101?), but never in a real design.
-
I would get the appropriate amount of D cells and put a LDO on that. Lantern batteries, or those yet bigger alkaline batteries, are even nicer. I had a 'everready' battery (supposedly a bad manufacturer) last 5 years in storage, without problems, but I took it apart to play with some carbon rods. Typically they say 3 year shelf life, the alkaline is 5+
The cost for using batteries is going to be astronomically lower then dealing with mains. I wonder how many batteries you would need to go through to justify the cost of that setup. :D . No offense to jim williams, but the price is ape shit considering it only measures 1 parameter in a big data sheet.
IMO spend the money on the enclosure and replace the battery in 5 years. I think a nice enclosure is important for this measurement, like steel.
-
For instance, this bad boy has 56 amp hours (alkaline)
https://web.archive.org/web/20090816013818/http://www.batteryspecialist.com/Merchant2/energizer/521.pdf (https://web.archive.org/web/20090816013818/http://www.batteryspecialist.com/Merchant2/energizer/521.pdf)
This is a 12V one with less amp hours (zinc)
https://web.archive.org/web/20090816014355/http://www.batteryspecialist.com/Merchant2/energizer/732.pdf (https://web.archive.org/web/20090816014355/http://www.batteryspecialist.com/Merchant2/energizer/732.pdf)
12V cell for 16$
https://www.bhphotovideo.com/c/product/1616349-REG/rayovac_926d_12v_general_purpose_screw_terminals.html?ap=y&smp=y&srsltid=Ad5pg_Hs-CVkuiy8rsrITEC_X30Pua7VAeGUYOy7-n05qinOu4iM4xSSJe0 (https://www.bhphotovideo.com/c/product/1616349-REG/rayovac_926d_12v_general_purpose_screw_terminals.html?ap=y&smp=y&srsltid=Ad5pg_Hs-CVkuiy8rsrITEC_X30Pua7VAeGUYOy7-n05qinOu4iM4xSSJe0)
-
No rechargeable option? LiFePo4 maybe? NiMh ?
-
Low noise power supply topic has been discussed many times, but it would be great to show such a practical proven design with easy to get parts, indeed. People do use 317 or 723 when talking low noise, or some special ADI/TI parts. The version with 317 as the pass element and an opamp with a low noise vref - I've seen it in the DS of an opamp (308 or 101?), but never in a real design.
I have used that configuration many times. I replace the JFET with a PNP bipolar, although a JFET or MOSFET should yield better performance by minimizing the load on the precision operational amplifier to maximize open loop gain. If a 317/337 is used, then care must be taken to keep the output of the operational amplifier within its output range, but for a 10 volt output, I last used a 7805 so that was not a problem.
Noise is limited by the performance of the reference, even with a low noise LM329. Load regulation is something like 1 microvolt per amp.
I recently asked someone who was doing design work back then about why the 108/308 was used. They confirmed that the 308 was selected because it was the first available part with low input offset voltage drift.
-
I'd take a look at Jim Williams' LT app note 124 (https://www.analog.com/media/en/technical-documentation/application-notes/an124f.pdf (https://www.analog.com/media/en/technical-documentation/application-notes/an124f.pdf)). Input PSU noise is only the start of your problems. I don't have a lot of experience, but I'd probably just go for something proven like a Keithley sourcemeter for the bench supply, then spend the rest of my time and effort on board-level regulator/reference design.
-
Walt Jung did a lot of work on low noise linear voltage regulator design. Regulators for High-Performance Audio (https://pearl-hifi.com/06_Lit_Archive/14_Books_Tech_Papers/Jung_W/Regs_for_High_Perf_Audio_1-4.pdf) Walt Jung Parts 1-4
I believe he settled on an LM317 as a pre-regulator and circuit like this (2000). It was in EDN as well, and been modded and audiophiles have their own tweaks. AD797 was a bit better for noise but I though worse for EMI susceptibility.
Some more history of the work: https://tangentsoft.com/elec/opamp-linreg.html (https://tangentsoft.com/elec/opamp-linreg.html)
-
..The version with 317 as the pass element and an opamp with a low noise vref - I've seen it in the DS of an opamp (308 or 101?), but never in a real design.
I have used that configuration many times. I replace the JFET with a PNP bipolar, although a JFET or MOSFET should yield better performance by minimizing the load on the precision operational amplifier to maximize open loop gain. If a 317/337 is used, then care must be taken to keep the output of the operational amplifier within its output range, but for a 10 volt output, I last used a 7805 so that was not a problem.
Noise is limited by the performance of the reference, even with a low noise LM329. Load regulation is something like 1 microvolt per amp.
I recently asked someone who was doing design work back then about why the 108/308 was used. They confirmed that the 308 was selected because it was the first available part with low input offset voltage drift.
The sim with the 7805, a p-jfet (it is picky, however), the LM308 (with its compensation inputs) and a high ESR capacitor works stable it seems, the issues start with 317+bipolar/mosfet+modern_opamp. Would be great to find a stable solution with easy to get parts..
-
No rechargeable option? LiFePo4 maybe? NiMh ?
Yeah you can but nothing really beats the simplicity of the alkaline cell. Done in 10 seconds without second thought kind of thing. Other batteries pose unique problems (very high short circuit current, cell bonding, protection circuits, charging circuits, charge ports). If you want to do away with that engineering, and get right to the critical design aspect, it will be faster, and also possibly cheaper. Lets just say you are not commiting anything at all for the alkaline cell and its usually OTC. Some people might appreciate that.. so I put it out there, because most people forgot those batteries exist besides for camping.
And of course if you just put a switch and a fuse, you can forget about it for 5 years, possibly much more, without having to do any kind of topping off. And even if you have some kind of battery, rechargable battery, its going to be harder to replace in the same form factor then one of those guys, in the future when you forgot the details but want to use a tool. Trouble shooting = trip to hardware store
And it has another benefit: if you want to go with rechargable for sure (and when future problems might show up) you can replace it with an alkaline, because those cells come in rechargable ones too., or switch to the rechargable one with a drop in replacement
https://www.amazon.com/6-volt-rechargeable-lantern-battery/s?k=6+volt+rechargeable+lantern+battery (https://www.amazon.com/6-volt-rechargeable-lantern-battery/s?k=6+volt+rechargeable+lantern+battery)
Not sure I trust the rechargable one that much though
And of course its like 5%+ a month discharge, vs a few percent per year with alkaline. I know which one I want for a esoteric amplifier I might forget about for half a decade until I need it again. ;D
-
Walt Jung did a lot of work on low noise linear voltage regulator design. Regulators for High-Performance Audio (https://pearl-hifi.com/06_Lit_Archive/14_Books_Tech_Papers/Jung_W/Regs_for_High_Perf_Audio_1-4.pdf) Walt Jung Parts 1-4
I believe he settled on an LM317 as a pre-regulator and circuit like this (2000). It was in EDN as well, and been modded and audiophiles have their own tweaks. AD797 was a bit better for noise but I though worse for EMI susceptibility.
Redraw the Jung's schematic and it is the same as the circuit posted by imo, with practically identical noise performance. Although not explicitly marked in the schematic, look carefully at where the bottom of the resistor divider and reference are connected.
-
The sim with the 7805, a p-jfet (it is picky, however), the LM308 (with its compensation inputs) and a high ESR capacitor works stable it seems, the issues start with 317+bipolar/mosfet+modern_opamp. Would be great to find a stable solution with easy to get parts..
The p-channel JFET should not be tricky at all. Nothing prevents the operational amplifier from driving the regulator directly except its output voltage range, however if the current is high, which will not be the case with a 317 which only requires an adjustment pin current of microamps, it will compromise the precision of a precision operational amplifier.
The regulator combined with a low noise operational amplifier, which is almost certainly faster than a 308, requires external compensation for stability. I add a lead network (two capacitors and one resistor) to the top of the resistor divider and a lag network (also two capacitors and one resistor) from the operational amplifier output to the inverting input, but not all of the parts are likely to be used. Values are mostly dependent on the regulator's frequency response combined with the output capacitance, so the output capacitor also significantly contributes to the frequency response but the recommendation for the regulator is suitable. Since the regulator's frequency response is poorly defined, I tune the frequency compensation empirically which is easy, and then the frequency response of the regulator can be calculated if necessary. Network analyzers have always been expensive luxuries for me.
I know of a much more sophisticated form of frequency compensation for difficult loads involving a feedback network with less than 90 degrees of phase lag, which increases the frequency margin, but it has not been required here.
Note that using a lower noise and faster operational amplifier will generally *not* lower the noise, because noise is dominated by the reference even when a low noise reference is used, so using a relatively noisy 308 is not a disadvantage. I last used a low noise LT1007/OP27, but in retrospect an LT1001/OP07 would have been just as good if not better, and easier to frequency compensate. A precision improved 308 part like the LT1008/LT1012/LT1097 might be better yet because of external compensation or overcompensation support. Of course these days there is no shortage of suitable precision parts with low flicker noise, low drift, and high gain.
High frequency performance is dominated by the output transistor or regulator and there is something to be gained there with careful design and a faster operational amplifier, but usually this is irrelevant.
-
Yeah you can but nothing really beats the simplicity of the alkaline cell. Done in 10 seconds without second thought kind of thing.
And I feel safe soldering wires directly to the ends of an alkaline cell. This is questionable for any lithium or lithium rechargeable cell without using tabs.
-
Yeah you can but nothing really beats the simplicity of the alkaline cell. Done in 10 seconds without second thought kind of thing.
And I feel safe soldering wires directly to the ends of an alkaline cell. This is questionable for any lithium or lithium rechargeable cell without using tabs.
Yep, same technique used back in 70s & 80s mentioned in post #1. Were used with custom test fixture to special select National transistor diff pair for sub-nanovolt/rtHz preamp work at very low frequencies. Later used with ultra-low phase noise oscillator measurements (1/f and low frequency noise matters even in microwave oscillators).
Still works well even today :-+
Best,
-
Boss found stash of special AA cells and used all powering Christmas LED fixtures :wtf:
Hope she doesn't see this post, as these are the special AAA cells we have hidden ;)
Hopefully these will last until we have to revert to using DuraHell or NeverReady :P
Best,
-
For low noise, I dislike (leaking alkaline batteries lol) and three-terminal regulators i.e LM317, 78xx because they seem to have gotten cheaper and noisier. That single-gauge Fairchild is the worst ever. As well, they are from an era of 100/120Hz ripple and too slow to deal with any SMPS ripple at their input. I would go without incorporating one.
AD797 was not the best for stability, had problems mentioned here Improved Positive/Negative Regulators (https://refsnregs.waltjung.org/Improved_PN_Regs.pdf) and he changed to AD825 or AD817.
Another EDN article he did that has good information in it: Low-Noise Power For Analog Circuits (https://pearl-hifi.com/06_Lit_Archive/15_Mfrs_Publications/04_Analog_Devices/Low_Noise_Power_For%20_Analog_Circuits.pdf)
-
I know of a much more sophisticated form of frequency compensation for difficult loads involving a feedback network with less than 90 degrees of phase lag, which increases the frequency margin, but it has not been required here.
Are you referring to fractional pole compensation? https://www.radioeng.cz/fulltexts/2016/16_02_0312_0320.pdf (https://www.radioeng.cz/fulltexts/2016/16_02_0312_0320.pdf)
I think its a really cool technique that is rarely discussed, despite the fact that I suspect most modern LDO regulators that are stable for wide ranges of capacitor ESR use it.
-
I know of a much more sophisticated form of frequency compensation for difficult loads involving a feedback network with less than 90 degrees of phase lag, which increases the frequency margin, but it has not been required here.
Are you referring to fractional pole compensation? https://www.radioeng.cz/fulltexts/2016/16_02_0312_0320.pdf (https://www.radioeng.cz/fulltexts/2016/16_02_0312_0320.pdf)
I think its a really cool technique that is rarely discussed, despite the fact that I suspect most modern LDO regulators that are stable for wide ranges of capacitor ESR use it.
Very interesting technique, thanks for posting.
Around 2000 in the early stages of developing RF, Microwave, MMW Systems on Chip (SoC) we required multiple on-chip LDOs for various functions. We utilized traditional PMOS output LDO designs which suffered the usual noise, and capacitor stability issues. One particular on-chip need was an ultra low noise LDO for biasing Voltage Controlled Oscillators (VCO) for use with Frequency Synthesis.
We developed a new type LDO that utilized a Super Source Follower type architecture which allowed unique noise decoupling to help bypass noise generated from the Bandgap Reference and other regulator elements, see patent 8692529.
https://patents.justia.com/patent/8692529 (https://patents.justia.com/patent/8692529)
Recall we were able to achieve a measured noise density 10~12nv/Hz @1KHz at the regulator output which included Bandgap Ref and all other internal regulator noise sources. At the time recall this was one of the lowest noise LDOs available, but surly there are better LDOs today. Back then a major microwave company became interested in a custom development LDO for their low Phase Noise VCOs as they were having to utilize custom discrete component LDOs at the time to meet the Phase Noise specification. Intel duplicated the concept (long story) for use within their processors for the on-chip DLL timing generators voltage regulators.
Anyway, the LDO Fractional Pole Compensation technique is indeed intriguing :-+
Best
-
I know of a much more sophisticated form of frequency compensation for difficult loads involving a feedback network with less than 90 degrees of phase lag, which increases the frequency margin, but it has not been required here.
Are you referring to fractional pole compensation? https://www.radioeng.cz/fulltexts/2016/16_02_0312_0320.pdf (https://www.radioeng.cz/fulltexts/2016/16_02_0312_0320.pdf)
That looks like the same idea although I do not know it under that name. Basically the capacitor in the commonly used integrator with 6dB/octave roll-off and 90 degrees of phase lag is replaced with an RC network resulting in an integrator with 3dB/octave roll-off and 45 degrees of phase lag, which adds 45 degrees of phase margin making the regulator much more stable.
I think its a really cool technique that is rarely discussed, despite the fact that I suspect most modern LDO regulators that are stable for wide ranges of capacitor ESR use it.
There is a much simpler solution for modern low dropout regulators. The collector/drain of the output transistor is tapped to produce an AC feedback signal. This produces what is effectively a resistance in series with the low ESR output capacitor at high frequencies, which adds phase lead to make the regulator stable. It is equivalent to adding a small series resistance to the low ESR capacitor to make a zero, but the series resistance is in a different spot.
Some of the early Linear Technology regulator datasheets discuss what is going on and how it is done.
-
I wonder whether a 723 + LT1021 combo would be "low noise" enough..
For smaller currents (up to 50mA let say) it may work without the transistor, imho..
PS: added 723+LM329 combo
-
Oh, 10 years old already!
I measured some stuff I had around:
< http://www.hoffmann-hochfrequenz.de/downloads/Noise_Measurements_On_Some_Laboratory_Power_Supplies.pdf (http://www.hoffmann-hochfrequenz.de/downloads/Noise_Measurements_On_Some_Laboratory_Power_Supplies.pdf) >
< http://www.hoffmann-hochfrequenz.de/downloads/NoiseMeasurementsOnChemicalBatteries.pdf (http://www.hoffmann-hochfrequenz.de/downloads/NoiseMeasurementsOnChemicalBatteries.pdf) >
Very low frequencies may be shown worse than they are because of the preamplifier.
When it counts, I use Lithium batteries.
Gerhard
-
I use 2DW232 ( metrology section ) as a reference, additionally trimmed for a minimum tempco
another option IR LED , as lowest noise source in LED, but 2DW232 at min. tempco really great. basic circuit opamp zener transistor ; zener fed from opamp , V output doubled .. 10.4V approx.
use capacitor multiplier to feed Ref. circuit and error opamp, 2-n order filter with 680 ohm resistors , with 5-10ma load - it does some filtering.
all ref / mgm circuit in enclosed shielded box , just pins for external connection, shielded variable resistors and shielded wire to connect sensitive points.
and it fail .... ringing when inductance load, ( maybe not :) ) 3 wirewound resistors ... oh well , seems it will be a new iteration ...