PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2020-09-25, 04:10:25
News: Forum TIP:
The SHOUT BOX deletes messages after 3 hours. It is NOT meant to have lengthy conversations in. Use the Chat feature instead.

Pages: [1]
Author Topic: Pulsed Inductive Circuits - Advice for Oscilloscope Power Measurements  (Read 4260 times)

Group: Administrator
Hero Member
*****

Posts: 2915
It's not as complicated as it may seem...
In my efforts to advise an individual that recently sought my advice on integral measurement, I am posting the following in hopes that it may be beneficial to those actually willing to embrace the valuable information that was offered. This will include partial emails, links, and a document I produced.  Note: all of the partial content is my own, and is not private nor confidential despite some of the email titles.

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Re: Private and confidential
Date: Sun, 19 Dec 2010 09:53:53 -0700

...I might also mention, that there is the issue of probe types and the physical layout/setup of your entire apparatus. These two factors had the greatest influence on the results Glen was getting with his setup. To my knowledge, he never implemented all of the corrections I proposed in my "Ainslie_Problems_Solutions02.pdf" document which I produced and made available to all those involved (posted etc), including you and Glen. I have attached it here again for your reference. You completely rejected all of it at the time, but perhaps there is a chance you will sincerely consider its contents this time.

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Re: private and confidential
Date: Sun, 19 Dec 2010 19:56:03 -0700

...my advice to you if you
are truly serious, is to hire a professional test engineer to test your
present apparatus. In doing so, accept nothing less than the use of a high
bandwidth differential voltage probe, a high bandwidth current probe, and a
high bandwidth quality scope. Have them produce a comprehensive test plan
first, then have a qualified independent review it to ensure you will get
what you need. Then you will have a set of proper measurements, and by
comparison to some properly obtained controls, you will be able to forever
put the question to rest.

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Re: private and confidential
Date: Mon, 20 Dec 2010 08:20:24 -0700

...If you can not find a company with the necessary expertise and equipment,
then it is up to you how you proceed. Keep in mind that the scopes are not
the problem, it is the probes and the setup of the apparatus that is of
paramount concern. If you go the route on your own, I would think that
somewhere down the line you will have to prove beyond any doubt whatsoever,
that the apparatus is truly OU. In order to prove it to "mainstream", the
testing will have to be performed with those probes, otherwise the results
will be dismissed. Rose, testing for power and energy this way, even with
the proper probes is tricky. The probes either have to be "matched", or the
scope has to have the facility to correct for skew between the probes. This
type of testing is not really in the realms of the "hobbyist", or even the
typical university student.

If the university could somehow get funding to buy a set of proper probes,
that would be a start. That would cost about $3k US. From that point,
someone with the right expertise in testing could guide the students to
performing the tests properly with these probes. Then you will have a much
better chance at convincing anyone that the results are accurate.

But step one, and I can't stress this enough, is to prove beyond a doubt
that the apparatus is OU, and as a minimum, the proper probes and testing
expertise need to be used for this testing. If there is no way you can get
this testing done as I suggest, then the ------ group may be your best
option.

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Re: private and confidential
Date: Tue, 21 Dec 2010 10:33:45 -0700

You need both suggested probes to do the testing properly, and yes grounding
is a big part of the issue, which is also why the probes are needed.
Standard passive probes such as the ones you are using are adequate for most
testing, but when it comes to measuring spiky signals that is a different
story. To complicate the issue further, it is ultra-critical that the two
suggested probes be "time matched" such that they do not skew the power
calculation. You are dealing with kickback pulses on the order of tens of
nano-seconds, and probes can easily delay signals 20 or more nano-seconds.
What that amounts to is an inaccurate power measurement.

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Re: private and confidential
Date: Tue, 21 Dec 2010 10:52:48 -0700

Read the document I sent you the other day. All that you need to know about
what I am saying, or trying to say, is in that document. It can not be made
any more clear.

I would suggest that you print that document out and suggest that the
students and/or professor over-seeing this project study and heed the advice
within it. It provides several graduated steps to improving the integrity of
these measurements.

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Accurate Instantaneous Power Measurements
Date: Tue, 21 Dec 2010 14:44:44 -0700

I would encourage you and your team to study this document from Lecroy very
closely, and note the many important points I have been trying to emphasize
to you:

http://www.zenixon.com/upfile/product/pdf00124.pdf


Also, these from Tektronix:

http://www.isotest.es/web/Soporte/Formacion/Notas%20de%20aplicacion/TEKTRONIX/DPO7000/TEK-POWER%20ANALYSIS%20PRIMER.pdf

http://www.scopeshop.de/(S(qlzwhd45rdgv4riyyit0oebi))/PRODUKTINFO/Oszilloskope/TDS5000B/File_1422/Sophisticated%20Power%20Loss%20Analysis.pdf

http://www.isotest.es/web/Soporte/Formacion/Notas%20de%20aplicacion/TEKTRONIX/DPO4000/MEDIDAS%20EN%20POWER%20SUPPLY%20DPO4000.pdf

http://www.imex.ie/files/u3/power_supply_measurement_and_analysis_with_the_mso_71349.pdf

And this article applies to your circuit as well:

http://www2.tek.com/cmswpt/tidetails.lotr?ct=TI&cs=afs&ci=14789&lc=EN

Quote
From: <poynt99@overunityresearch.com>
To: "Rosemary Ainslie" <ainslie@mweb.co.za>
Subject: Re: WHAT IS THE ACTUAL GAME HERE?
Date: Tue, 21 Dec 2010 21:49:10 -0700

As explained in those articles and by myself, we are not referring to
marginal errors where 0.5W may be missing here or there...it's more to the
issue of the measurements can be completely erroneous, even in favour of
more power out than in. I've tried my best to make you aware of why and how
this can happen. You don't have to trust me, and you don't have to trust the
Tektronix and Lecroy engineers who design these DSO's and expensive probes,
AND whom not the least of which, are experts in measurement. But it will be
to your disadvantage if you don't.

See also the attached document pertaining to properly measuring the power and or energy for a typical pulsed inductive circuit.

.99

   

Group: Administrator
Hero Member
*****

Posts: 2915
It's not as complicated as it may seem...
As an alternative approach if a current probe is not available, try using a differential voltage probe in combination with a non-inductive current sensing resistor (CSR).

As pointed out in the document, even this approach can improve the accuracy of the power measurement tremendously. The one important consideration however, is to ensure that the probe tips of the differential probe are placed as close to the CSR body as possible. Any wire lead existing between the two probes introduces stray inductance which will skew the current measurement, thus resulting in an erroneous instantaneous power computation. A high voltage differential probe will not be necessary for this as the voltage levels across a CSR are normally quite low.

.99
   

Group: Administrator
Hero Member
*****

Posts: 2915
It's not as complicated as it may seem...
I see that TK has made note of my above post and produced an excellent video clearly demonstrating the detrimental effects of not having the probe tips as close as possible to the resistor body.

He is not using a differential probe, but the same rule applies to single-ended passive probes as well. Well done TK.  O0

[youtube]http://www.youtube.com/watch?v=KWDfrzBIxoQ[/youtube]

.99
   
Pages: [1]
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2020-09-25, 04:10:25