Author Topic: Affordable <200MHz PDN analysis / impedance measurement hardware?  (Read 7577 times)

0 Members and 2 Guests are viewing this topic.

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #25 on: November 12, 2024, 09:01:21 am »
I ended up designing a little adapter board for the Analog Discovery series!

Here's the board measuring a low impedance capacitor array:

Hi,
Could you please share the schematic?

Available on my github page: https://github.com/remidavidkoster/Redako-PDNAnalyzer
 
The following users thanked this post: egonotto, Kean, ch_scr

Offline axb

  • Contributor
  • Posts: 14
  • Country: pl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #26 on: November 12, 2024, 09:41:57 am »
Super! Thanks!
I only see one uncertainty. In my WaveForms (AD3) I don't have the selectable configuration you indicate, I only have DUT-RES-GND.
 
The following users thanked this post: egonotto

Offline mawyatt

  • Super Contributor
  • ***
  • Posts: 4120
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #27 on: November 12, 2024, 02:41:01 pm »
Very impressive results with the low Z range covered :-+

Curious why you selected Tantalums rather than Polymer or Aluminum Electrolytic Capacitors for the coupling/decoupling, size maybe?

Best
Curiosity killed the cat, also depleted my wallet!
~Wyatt Labs by Mike~
 
The following users thanked this post: egonotto

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #28 on: November 12, 2024, 02:49:37 pm »
Very impressive results with the low Z range covered :-+

Curious why you selected Tantalums rather than Polymer or Aluminum Electrolytic Capacitors for the coupling/decoupling, size maybe?

Best

Hey Mawyatt,

Thanks! Pretty stoked with it myself as well.

Indeed size, and with that inductance. The shunt is 500mΩ. If I would use a capacitor with an equivalent series inductance of 10nH I'd see an additional impedance of 1.57Ω in series with the rest of the circuit at 25MHz, limiting the current I'd be able to get through with a factor of 4, and with that reducing the voltage I'd read back on the second sense channel with a similar factor of 4.

Keeping the capacitor and its inductance as small as possible allows most current at high frequencies, keeping my sense voltages as high as possible there.
 

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #29 on: November 12, 2024, 02:50:38 pm »
Super! Thanks!
I only see one uncertainty. In my WaveForms (AD3) I don't have the selectable configuration you indicate, I only have DUT-RES-GND.

I think you should be able to select that one, and switch the scope channels around somewhere else. Then you effectively have the same configuration.
 

Offline mawyatt

  • Super Contributor
  • ***
  • Posts: 4120
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #30 on: November 12, 2024, 03:35:46 pm »
Very impressive results with the low Z range covered :-+

Curious why you selected Tantalums rather than Polymer or Aluminum Electrolytic Capacitors for the coupling/decoupling, size maybe?

Best

Hey Mawyatt,

Thanks! Pretty stoked with it myself as well.

Indeed size, and with that inductance. The shunt is 500mΩ. If I would use a capacitor with an equivalent series inductance of 10nH I'd see an additional impedance of 1.57Ω in series with the rest of the circuit at 25MHz, limiting the current I'd be able to get through with a factor of 4, and with that reducing the voltage I'd read back on the second sense channel with a similar factor of 4.

Keeping the capacitor and its inductance as small as possible allows most current at high frequencies, keeping my sense voltages as high as possible there.

WRT to the Coupling Capacitance and ESL/ESR. Have you considered using a 0.1uF C0G/NP0 ceramic in parallel with a larger polarized cap? These have a low ESL and ESR, we've done such with our DC Bias Adapter for use with Bench LCR Meters and seems to work well.

This is tempting to pull out our AD2 and put it to use ::)

Best
Curiosity killed the cat, also depleted my wallet!
~Wyatt Labs by Mike~
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #31 on: November 12, 2024, 03:44:21 pm »
I am in the process of updating Solver's user manual and repeated the PDN measurements with my H4 using the latest firmware available.  I had no lockups or glitches in the data. 

I had damaged my H4 (suspect ESD) and when repairing it, I had added a few additional caps which greatly improved the noise floor at higher frequencies.     Still, it is not good enough to resolve 100uOhms.   

Link to troubleshooting, mods and performance gains of my H4.
https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg5695349/#msg5695349

I also have the newly release LiteVNA hardware.   I was very much surprised that it performs slightly better then my other low cost VNAs at these lower impedance's, however it is still not able to resolve 100uOhms. 

That said,  typically when I run into problems with emissions, they are in the several MHz on up past GHz.   Assuming you are wanting to measure the PDNs stability for emissions, how do you test it at these higher frequencies? Or is it not a concern? 

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #32 on: November 12, 2024, 06:09:22 pm »
Have you considered using a 0.1uF C0G/NP0 ceramic in parallel with a larger polarized cap? These have a low ESL and ESR, we've done such with our DC Bias Adapter for use with Bench LCR Meters and seems to work well.

Best

I haven't optimized that part too much - the little coaxial cables I solder to the board have quite some inductance as well, so I think the effect on the total wouldn't be too significant.

I took a look at your DC Bias adapter posts - nice stuff too. :)
 

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #33 on: November 12, 2024, 06:42:32 pm »
That said, typically when I run into problems with emissions, they are in the several MHz on up past GHz. Assuming you are wanting to measure the PDNs stability for emissions, how do you test it at these higher frequencies? Or is it not a concern?

Nice writeup on the H4 mods!

With the measurements I did so far (of single capacitors and small PCB PDNs) I've been able to measure high enough in frequency to get down to the package- / measurement setup inductances, which is more than enough for me (I mainly have audio clients at the moment), but indeed not getting to the point of plane resonances and the like. I believe because of the inductance involved the impedances here are typically a lot higher (>100mΩ), so getting out the VNA for that would probably be fine? What do you think?
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #34 on: November 12, 2024, 07:23:13 pm »
That said, typically when I run into problems with emissions, they are in the several MHz on up past GHz. Assuming you are wanting to measure the PDNs stability for emissions, how do you test it at these higher frequencies? Or is it not a concern?

Nice writeup on the H4 mods!

With the measurements I did so far (of single capacitors and small PCB PDNs) I've been able to measure high enough in frequency to get down to the package- / measurement setup inductances, which is more than enough for me (I mainly have audio clients at the moment), but indeed not getting to the point of plane resonances and the like. I believe because of the inductance involved the impedances here are typically a lot higher (>100mΩ), so getting out the VNA for that would probably be fine? What do you think?

Without knowing if you solved the ground loop problem, its hard to say.   Also, our goals for making these measurements are not the same.  You may be better off with what you have for now.   

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #35 on: November 12, 2024, 08:25:09 pm »

Without knowing if you solved the ground loop problem, its hard to say.   Also, our goals for making these measurements are not the same.  You may be better off with what you have for now.

Having a fully differential measurement I have no shared ground path and thus no ground loops - the same as when using a for example a J2113A (Semi-Floating Differential Amplifier) with a VNA. Just like a Four-Wire Kelvin measurement, where a VNA is more like a Three-wire measurement.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #36 on: November 12, 2024, 08:50:37 pm »

Without knowing if you solved the ground loop problem, its hard to say.   Also, our goals for making these measurements are not the same.  You may be better off with what you have for now.

Having a fully differential measurement I have no shared ground path and thus no ground loops - the same as when using a for example a J2113A (Semi-Floating Differential Amplifier) with a VNA. Just like a Four-Wire Kelvin measurement, where a VNA is more like a Three-wire measurement.

So your plan is to use that same basic circuit you are showing with your Analog Discovery with your H4?   I don't see a problem with breaking the VNA's ground this way.   

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #37 on: November 12, 2024, 09:01:14 pm »
So your plan is to use that same basic circuit you are showing with your Analog Discovery with your H4?   I don't see a problem with breaking the VNA's ground this way.

Sorry - I completely misunderstood what you said. It’s been a long day. Disregard what I said.

I haven’t solved it yet, but thought it might be accurate enough using it above 100mOhm to cover the range above 25Mhz.

I do still plan to make a differential amplifier for it with the IC nctnico used in his 1.4Ghz differential probe, but haven’t gotten to it yet.
 

Offline Smokey

  • Super Contributor
  • ***
  • Posts: 3046
  • Country: us
  • Not An Expert
Re: Affordable <200MHz PDN analysis / impedance measurement hardware? [Solved!]
« Reply #38 on: November 12, 2024, 09:10:50 pm »
...
Here's the board measuring a low impedance capacitor array:

...

what is that capacitor board???  I gots to know!  purpose, total capacitance, how the hell did they place/solder all those caps when they are damn near touching??
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #39 on: November 12, 2024, 09:29:08 pm »
So your plan is to use that same basic circuit you are showing with your Analog Discovery with your H4?   I don't see a problem with breaking the VNA's ground this way.

Sorry - I completely misunderstood what you said. It’s been a long day. Disregard what I said.

I haven’t solved it yet, but thought it might be accurate enough using it above 100mOhm to cover the range above 25Mhz.

I do still plan to make a differential amplifier for it with the IC nctnico used in his 1.4Ghz differential probe, but haven’t gotten to it yet.

In the manual, I show measuring my standards using my old HP network analyzer at 2.5kHz.  The 0.1 ohm standard had 2.74% error without the common mode transformer.  With it, 0.54%.  I could repeat this at higher frequencies but it will only get worse.   I could also attempt it with my H4 with / without transformer.    I doubt it would meet your requirements with a direct connection.       

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware? [Solved!]
« Reply #40 on: November 12, 2024, 09:32:03 pm »
...
Here's the board measuring a low impedance capacitor array:

...

what is that capacitor board???  I gots to know!  purpose, total capacitance, how the hell did they place/solder all those caps when they are damn near touching??

Hehe, I wondered when someone would scratch their head about those. I had them made a while ago for both fun (they spot weld) and for those occasions where someone says “maybe we can solve it by adding more capacitors”. There’s a total of 6960 10V 10uF 0603 capacitors on there - cheapest I could find per uF - adding up to around 70mF, with a negligible ESR (limited by the 35oz copper planes) and an ESL limited by the wires you solder on there. Works great at eradicating supply ripple if it’s within the right bandwidth. I also have some stacked together with bismuth solder. :)

The alternating rows of capacitors are flipped in polarity, so the ends nearly contacting other ends are all the same net. Placement is done with tiny vacuum nozzles, soldering with solder paste and a reflow oven.
 

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #41 on: November 12, 2024, 09:36:30 pm »
In the manual, I show measuring my standards using my old HP network analyzer at 2.5kHz.  The 0.1 ohm standard had 2.74% error without the common mode transformer.  With it, 0.54%.  I could repeat this at higher frequencies but it will only get worse.   I could also attempt it with my H4 with / without transformer.    I doubt it would meet your requirements with a direct connection.       

Hmm, shit, I blindly hoped it would stay the same regardless of frequency. I’ll start cooking up a board for that differential amplifier.
 

Online tszaboo

  • Super Contributor
  • ***
  • Posts: 8218
  • Country: nl
  • Current job: ATEX product design
Re: Affordable <200MHz PDN analysis / impedance measurement hardware? [Solved!]
« Reply #42 on: November 12, 2024, 09:59:02 pm »
...
Here's the board measuring a low impedance capacitor array:

...

what is that capacitor board???  I gots to know!  purpose, total capacitance, how the hell did they place/solder all those caps when they are damn near touching??

Hehe, I wondered when someone would scratch their head about those. I had them made a while ago for both fun (they spot weld) and for those occasions where someone says “maybe we can solve it by adding more capacitors”. There’s a total of 6960 10V 10uF 0603 capacitors on there - cheapest I could find per uF - adding up to around 70mF, with a negligible ESR (limited by the 35oz copper planes) and an ESL limited by the wires you solder on there. Works great at eradicating supply ripple if it’s within the right bandwidth. I also have some stacked together with bismuth solder. :)

The alternating rows of capacitors are flipped in polarity, so the ends nearly contacting other ends are all the same net. Placement is done with tiny vacuum nozzles, soldering with solder paste and a reflow oven.
I think everyone had the idea of just using a lot of capacitors, but you took this to a new level.

Your circuit is quite interesting. Any reason you decided to AC couple the signals instead of accepting the small DC offsets?
 

Offline rhb

  • Super Contributor
  • ***
  • Posts: 3518
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #43 on: November 12, 2024, 10:20:38 pm »
An HPAK 4395A is reasonably priced and had a very long life, so plentiful.   I’m using one to develop EMI filters to suppress conducted noise from my LED shop lighting. 

It will really get a workout when I start building the screen room feeds.

I really can’t see how you can improve on the price - performance.   I love it.  Best possible tool for the frequency range of interest.  It’s both a VNA and an SA with 1 Hz resolution.  It covers  10 Hz to 500 MHz.  For HF work it’s the king.

Designing appropriate fixtures and calibration standards for testing boards won’t be easy.  You’ll need machine tool access.  Helps a lot if you design in test points.  Low impedance fixtures are likely to be “interesting” to make.  But then the challenge is the whole point for a lot of projects.

Have Fun!
Reg
 

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware? [Solved!]
« Reply #44 on: November 12, 2024, 10:23:24 pm »
Your circuit is quite interesting. Any reason you decided to AC couple the signals instead of accepting the small DC offsets?

Thanks! When for example measuring the output impedance of a buck converter under load the DC bias will be equal to the output voltage, so for measurements like this being able to handle high DC offsets is essential
 

Offline inevitableavoidanceTopic starter

  • Contributor
  • Posts: 33
  • Country: nl
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #45 on: November 12, 2024, 10:31:11 pm »
An HPAK 4395A is reasonably priced and had a very long life, so plentiful.   I’m using one to develop EMI filters to suppress conducted noise from my LED shop lighting. 

It will really get a workout when I start building the screen room feeds.

I really can’t see how you can improve on the price - performance.   I love it.  Best possible tool for the frequency range of interest.  It’s both a VNA and an SA with 1 Hz resolution.  It covers  10 Hz to 500 MHz.  For HF work it’s the king.

Designing appropriate fixtures and calibration standards for testing boards won’t be easy.  You’ll need machine tool access.  Helps a lot if you design in test points.  Low impedance fixtures are likely to be “interesting” to make.  But then the challenge is the whole point for a lot of projects.

Have Fun!
Reg

That does look like the absolute perfect measurement range - a bit out of my initial budget but I’ll definitely keep an eye out for a good second hand deal. Test fixtures and connections are quite a world of their own I have indeed found. So far I’ve had most luck simply stripping and soldering on microcoaxial cables. I wish someone would make and standardise a miniature “kelvin” RF connector for this
 

Online tszaboo

  • Super Contributor
  • ***
  • Posts: 8218
  • Country: nl
  • Current job: ATEX product design
Re: Affordable <200MHz PDN analysis / impedance measurement hardware? [Solved!]
« Reply #46 on: November 12, 2024, 10:43:45 pm »
Your circuit is quite interesting. Any reason you decided to AC couple the signals instead of accepting the small DC offsets?

Thanks! When for example measuring the output impedance of a buck converter under load the DC bias will be equal to the output voltage, so for measurements like this being able to handle high DC offsets is essential
Ah, you want to measure it while on, not like with the VNAs. I believe a power rail probe is more suited for a measurement like that. I made a power rail probe, goes to 2GHz, no gain though. But with the small noise coming from it, you should be able to directly measure a few mv signals. And gain can be added by any 50 Ohm input/output amplifier.

Now, I'm not trying to sell you my probe, just want to give you some pointers, if you are interested.
 

Offline rhb

  • Super Contributor
  • ***
  • Posts: 3518
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #47 on: November 12, 2024, 11:12:51 pm »
Don’t underestimate a DSO with memory and a decent sig gen.   Suck the data into Octave for analysis.  The really hard part is a low frequency directional bridge.  I suggest reading up on slotted lines and how the microwave pioneers coped.  Measurements at multiple points on a delay line should handle direction separation.

If time is more plentiful than money, even the 20 kPt Hanmatek DSOs made by Owon and lots of measurements will do the job.  Use what you have.

Most of my life I was not willing to spend money on what were “toys”.  At 71 I now spend it. But I learned a lot working around no budget.  So I urge doing that very strongly.  It reinforces being able to work from first principles.  That was ALWAYS valuable.

Have Fun!
Reg

BTW Join the T&M list on groups.io.  The eBay prices are rather stiff.  And be patient.
« Last Edit: November 12, 2024, 11:16:33 pm by rhb »
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12280
  • Country: us
Re: Affordable <200MHz PDN analysis / impedance measurement hardware? [Solved!]
« Reply #48 on: November 12, 2024, 11:29:52 pm »
Ah, you want to measure it while on, not like with the VNAs.

?

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: Affordable <200MHz PDN analysis / impedance measurement hardware?
« Reply #49 on: November 13, 2024, 12:30:59 am »
Don’t underestimate a DSO with memory and a decent sig gen.   Suck the data into Octave for analysis.  The really hard part is a low frequency directional bridge.
Why would you want to use a directional bridge? For PDN you use a 2 port measurement AFAIK. Push AC current into the DUT from the transmitter port (generator) and measure the resulting AC voltage across the DUT.
« Last Edit: November 13, 2024, 12:43:08 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: egonotto


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf