Products > Test Equipment
DIY Transformer for use with Bode Plots.
<< < (9/14) > >>
joeqsmith:
Attempted to use the LiteVNA to looking at the last transformers gain.   The LiteVNA is spec'ed to work down to 50kHz.   Things really fall apart below 10kHz.   
mawyatt:

--- Quote from: joeqsmith on May 28, 2022, 11:55:03 pm ---I marked up Jay_Diddy_B's drawing I linked above.  Obviously in the first two circuits changing the location of CH1 is going to have a major effect what the low frequency response looks like when we measure our transformer.   Consider as we approach DC the primary will start to load the output.  Because we are looking at a ratio of ch1 to 2, it can make the response seem flat.   

I'm not a fan of the added resistor he shows but at DC, hey..   Anyway,  the third circuit, I change the scopes inputs to 50 ohms on both channels and include a splitter.   

I suspect the initial setup is wrong as Jay pointed out.  Calculating the rolloff we won't be anywhere near 10Hz as expected.  Maybe it adds to the confusion.

--- End quote ---

The original setup is not wrong, it's exactly as stated multiple times and illustrated in the Bode Plot video shown above. The Bode measurement with the Siglent DSO Ch1 is Across the Transformer Primary and Ch2 across the secondary with input impedance set to 50 ohms. Since the intended use and thread title of the transformer as stated is for Bode use, and mostly as an Injection Transformer for Closed Loop measurements, this method as stated and used partially removes the transformer (and source) characteristics from the measurement and revels the effective use of the transformer for Bode use.

This is NOT directly intended (although can be used) for a controlled impedance type use like 50 ohm RF, but generic Bode use where the effects of the input source are partially removed by the input sampling location. The intended case is for Closed Loop measurements the input sampling is actually moved to the transformer secondary side, the transformer is for isolation with the DUT, and one secondary winding wire goes to Ch1 as Input and the other secondary winding wire as Output goes to Ch2, the secondary creates a "floating injection source" inserted in series with the negative feedback (with restrictions) for the DUT. Since the Bode plots are of complex type (Magnitude and phase) the result loop response of the DUT is simply Vout/Vin in magnitude and phase. Note the effects of the signal source, transformer, cables are completely removed (ideally) from the result and why this technique is so powerful when applied properly!!

Here's some previous quotes in this thread answering your questions regarding the transformer intended use, and what the plots show.

Our intent was to show the CM Filter cores in a DIY configuration utilizing the wire from the CM Filter can be useful for Bode type Closed Loop Measurements within a 1MHz frequency range,

Here's what I think you are asking for, these are with 50 ohm source drive, transformer input measured with DSO Hi Z (1M) and output terminated with 50 ohms (DSO)


The low frequency impedance to the transformer is low, however the input sampling is across the transformer primary so this compensates some for the transformer low frequency characteristics.

Of course the intrinsic transformer doesn't have a basic 10Hz 3dB corner, just look at the primary inductance measurements of ~8mH which implies an impedance of 1/2 ohm at 10Hz!!! This is not to say the core transformer low end isn't important for Close Loop Injection use, the transformer must inject a signal into the DUT and this injection level falls off at the high and low frequency end due to transformer and source characteristics, as well as other effects. In the Closed Loop Bode measurement the loop gain of the DUT comes into play, and usually the loop gain has a general low pass type charteristic. So a rising DUT loop gain as frequency decreases causes the DSO Ch1 signal to be small as the negative feedback attempts to "null out" the injected signal which appears to the Closed Loop System as an error signal. The DSO will increase the input sensitivity to help compensate for such, but eventually reaches a sensitivity which can no longer "pull out" the signal from the noise, and the Bode response degrades. The same happens at the high end where the transformer rolls off, thus the injected signal rolls off. However, the DUT loop gain is also falling off which partially helps with DSO detecting the signals since less of the injected signal is "nulled out" by the negative feedback loop gain.

We realize this is a rather complex subject and likely difficult to get ones "arms around", especially with all the nuances involved in proper setup, use, and measurement understanding. Please spend some time studying the mentioned video (and other related papers on Bode Plots and Close Loop Measurement Techniques), this is an excellent resource for getting an understanding of Closed Loop Bode measurements, and even illustrates how this method can be utilized to measure the very complex non-linear nature of SMPS even tho Bode is a linear type function. 

Awhile back, well before this thread, some folks downplayed the Bode capability of these DSOs. Likely from a lack of knowledge and/or understanding just how useful this technique is when properly applied with the DSO. PicoScope, Siglent have this capability (we know and use both), Keysight and others also.

Anyway, hope this helps clarify what this thread was intended to illustrate, and how useful these DSOs are for Closed Loop Bode measurements with the simple addition of a repurposed inexpensive Common Mode Filter core for Bode Injection Transformer usage.

Best,
mawyatt:

--- Quote from: joeqsmith on May 29, 2022, 01:59:31 am ---Attempted to use the LiteVNA to looking at the last transformers gain.   The LiteVNA is spec'ed to work down to 50kHz.   Things really fall apart below 10kHz.

--- End quote ---

Not surprised at this result at lower frequencies! Pretty much useless at frequencies where Closed Loop Bode plots are useful, not to mention dealing with 50 ohm inputs where the DUT must not be perturbed by the measurement sensing probes.

This is where these newer DSO really shine, not long ago to do these types of Bode measurements one required a dedicated instrument, now it's built-in :-+

Best,
joeqsmith:
Indeed it does help.   You're looking at it from an application view where I am just comparing the performance of two transformers in a 50 ohm system.   I thought the goal of your first post was to show the transfer function of the transformer in a 50 ohm system which was partly why you initially had added these 50 ohm parts.    Indeed we can calculate the low roll off.  As your link to Jay's talks about, what you showed seemed too good to be true.   Of course, we can't see see the low response with how you are measuring it.   It's just my misunderstanding of what you were trying to present.

The data I have shown is with the transformers in a 50 ohm system.  I was curious how flat I could get the low frequency response.  I assumed that the person Jay linked to wanted that 0.1dB flatness because they have no way to compensate for the error and would not know what was being injected.     

Shown is the same core on the original NanoVNA.  These VNAs have better performance at lower frequencies but they are not going to get you near 10Hz.  Agree that its not useful for your application and more an FYI.       

I haven't spent much time looking into Siglent products.  After there was that whole thread about them going after eBay customers, I never considered them.  Eventually we procured an arb from them where I work just to try get a feel of the quality.  That was my first experience with the brand.   Interesting to see the built in Bode plot.  To do this today, I would use a separate scope, generator and PC.   

mawyatt:

--- Quote from: joeqsmith on May 29, 2022, 05:16:25 pm ---
I haven't spent much time looking into Siglent products.  After there was that whole thread about them going after eBay customers, I never considered them.  Eventually we procured an arb from them where I work just to try get a feel of the quality.  That was my first experience with the brand.   Interesting to see the built in Bode plot.  To do this today, I would use a separate scope, generator and PC.   

--- End quote ---

Don't know about the eBay issue with Siglent? What happened?

Frankly, we never expected the Siglent SDS2104X Plus to be as good as has demonstrated over and over in our use. This was our 1st venture into the new DSO/MSO arena with our personal $ funds, before retiring everything was HPAK Tek, LeCroy and R&S, so very selective and cautious!!

After much review here, learning and "selective listening" to those in the know, we pulled the trigger and acquired one (have 2 now). Even after 2 years of use it still amazes in performance and overall capability. May sound like a Siglent fanboy, but that's not the case as Siglent has some products that we're not particularly fond of, but generally their stuff is pretty good and good value, the SDS2104X Plus being an absolute outstanding value for serious use IMO. The front end design, noise level, offset DC range, waveform accuracy, computational accuracy, Display, FFT, Bode, Lan & Web interface and so on are all really good, and using a wireless mouse is icing on the cake :-+

The PicoScope is good, we needed this for the 16bit ADC precision waveform evaluation and it served the purpose well. It's also quite capable as a low frequency DSO, but never been a fan of laptop instruments, so not used very often.

Best,
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod