Author Topic: Why does mains transient tester only get up to 1600V when set for 2kV?  (Read 769 times)

0 Members and 1 Guest are viewing this topic.

Offline FaringdonTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 2124
  • Country: gb
Hi,
The attached shows a burst mode transient voltage applied to our mains connected equipment.
This comes from a Schaffner BEST  burst mode tester.
I set it to 2kV Peak...but as can be seen, its only 1600Vpk. (the zero ref for the "yellow"
channel is at -1600Vdc).
Why did it not get up to 2kV?
I measured it with a 100:1 probe connected from true mains earth (not the earth coming out of the Schaffner)
and Line. (Line that  comes out of the Schaffner)
The Schaffner was set up as follows....
_2kv peak
_No of burst pulses = 1
_Phase angle of burst = 90 degrees
_Transient coupled to Line only.

So why did it only get up to 1600V?
Was it due to the MOV etc?
« Last Edit: July 27, 2024, 09:29:58 pm by Faringdon »
'Perfection' is the enemy of 'perfectly satisfactory'
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
That's some beginner stuff there.  The generators are spec'ed for open circuit voltage and short circuit current waveforms (combo).    They have a specified source impedance as well.  Drive it into a dead short and the voltage will be less.   

You really would be better off renting some time at a lab. 
 
The following users thanked this post: Kean, Faringdon

Offline FaringdonTopic starter

  • Super Contributor
  • ***
  • !
  • Posts: 2124
  • Country: gb
Quote
The generators are spec'ed for open circuit voltage and short circuit current waveforms (combo).
Thanks,  say we have a Boost PFC'd power supply with 800uF of electrolytics after the booster. Then i take it that we can set 2kV, but will get way, way under that.
As you kindky discuss, it will clamp at the current limit. I must admit, i didnt see from the manual what that was. I think those pages are missing.
(theres no manual online for the Schaffner BEST)
Thanks but We cant afford to pay for lab time.
'Perfection' is the enemy of 'perfectly satisfactory'
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
Quote
(theres no manual online for the Schaffner BEST)

As you stated in your previous thread:
https://www.eevblog.com/forum/testgear/schaffner-best-transient-tester/

Quote
Thanks but We cant afford to pay for lab time.
Can you afford not to?

I would just run it open circuit, and then with some resistive loads until you at least understand some of the basics.  With it being so old an unsupported, are you sure it even meets the standards you are required to adhere to?  Or is this more just for fun and you really are not trying to meet any of the current standards?
 
The following users thanked this post: Kean, Faringdon

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7265
  • Country: ca
I am familiar with Teseq ones, super easy to use so techs can do the testing. It's a very high cost of ownership with commercial mains transient generators.
A new standard rolls out, or an update to a standard and you have to send it in for hardware mods- or the unit just goes obsolete. Steady churn of these units on the surplus market.
Calibration is a constant cash drain, for test-lab requirements. New, they are the price of a car.
That's why I wanted joeq to give details about the insides of his homebrew one ;)

If output voltage is low, make sure the unit is capable of that in the first place. It could be for older (standards).
It could need repairs, have some damage and only be making 1.6kV not 2kV, or have high output resistance. Or there is an arc in the DUT.
I'd test it with a simple resisitive load to make sure it's working properly.

Most important is safe test leads and the blast shield, on an insulated bench. You have to know what you are doing. We wouldn't want a dead Treez.
 
The following users thanked this post: Kean, Faringdon

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9787
  • Country: gb
... We wouldn't want a dead Treez.

A nasty fright to make you actually listen and think for yourself however.  >:D
Best Regards, Chris
 
The following users thanked this post: Faringdon

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7265
  • Country: ca
... We wouldn't want a dead Treez.

A nasty fright to make you actually listen and think for yourself however.  >:D

If you buy used test equipment and the safety hardware is missing, it can accelerate Darwin's process if self-reflection and intellect is also absent from the kit.
Icarus mentioned in the manual's missing pages. He thought he was pretty good at SMPS design, a bit cheap on the bee's wax though.
 
The following users thanked this post: Faringdon

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
Calibration is a constant cash drain, for test-lab requirements. New, they are the price of a car.
That's why I wanted joeq to give details about the insides of his homebrew one ;)

Sorry but I don't follow. People here like automotive analogies.  If a new McLaren was a quarter mil and I show a home made wind up toy, how does showing how the wind up toy was designed help?   The design constraints between a generator designed for the IEC standard and what I came up with to test multimeters is completely different.   I've been asked about transient testing AC mains products.  It shows just how far off some of the thoughts are about what was shown.  It doesn't seem to matter how much I try to educate viewers, people see and hear what they want.  In this case, OP I assume is wanting to test using an actual combo generator to some IEC standard.

 
Quote
Most important is safe test leads and the blast shield, on an insulated bench.
For mains products like the OP is asking about, the leads may very well be the supplied line cord.  I would imagine in many cases, there is no protective shield used.   Do use some sort of shield where you work?  If so, and you are able to discuss, may be worth a new topic. 


 
The following users thanked this post: Faringdon

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7265
  • Country: ca
Calibration is a constant cash drain, for test-lab requirements. New, they are the price of a car.
That's why I wanted joeq to give details about the insides of his homebrew one ;)

Sorry but I don't follow. People here like automotive analogies.  If a new McLaren was a quarter mil and I show a home made wind up toy, how does showing how the wind up toy was designed help?   The design constraints between a generator designed for the IEC standard and what I came up with to test multimeters is completely different.   I've been asked about transient testing AC mains products.  It shows just how far off some of the thoughts are about what was shown.  It doesn't seem to matter how much I try to educate viewers, people see and hear what they want.  In this case, OP I assume is wanting to test using an actual combo generator to some IEC standard.

I'd asked what component you used for the power switch in your transient generator, curious what could take that much voltage and current. Not for formal McLaren kind of building or scrutiny. It was simple wondering, towards that thread getting around using a relay.


Most important is safe test leads and the blast shield, on an insulated bench.
For mains products like the OP is asking about, the leads may very well be the supplied line cord.  I would imagine in many cases, there is no protective shield used.   Do use some sort of shield where you work?  If so, and you are able to discuss, may be worth a new topic.

It's a 1/4" thick plexiglass U-cover that keeps sparks, molten metal, component bodies from flying across the lab. It might have been made by another engineer, or is a Teseq accessory, not sure. TO-247 explodes and epoxy flies across the lab. Eye protection is good but getting beaned is still unpleasant.
When you have noobs and people walking around, pretty much mandatory in a corporate lab environment.

The cover I suppose is similar to the need for your multimeter test enclosure with the door and window, except this was sized for larger exposed PC boards under test.
 
The following users thanked this post: Faringdon

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
It's a 1/4" thick plexiglass U-cover that keeps sparks, molten metal, component bodies from flying across the lab. It might have been made by another engineer, or is a Teseq accessory, not sure. TO-247 explodes and epoxy flies across the lab. Eye protection is good but getting beaned is still unpleasant.
When you have noobs and people walking around, pretty much mandatory in a corporate lab environment.

When I worked in automotive, we would run Load Dump tests as part of our conducted transients.  Most of the time we tested the complete assembly and did not use any type of blast shield.  If we were testing bare boards, sometimes we would use a cardboard box to catch potential shrapnel.   When I changed my career to the world of AC mains where we use Burst and Surge, we pretty much only test with the complete assembly and no additional shield.    The people walking around isn't much of a concern as the labs have for the most part been isolated (separate building).  The only people there are the ones involved with the testing.  That said, some of us will test our circuits open framed.  I would say we are hit and miss if additional shielding is used or not.  Rather than a corporate policy, it seems to fall on the individuals running the tests.


Quote
The cover I suppose is similar to the need for your multimeter test enclosure with the door and window, except this was sized for larger exposed PC boards under test.

That setup works well for testing bare boards that I know have clearance issues or glass fuses.   :-DD   Even then, it's only used when I have that half cycle low voltage line simulator attached.   At the sub 20 Joules I normally test at, there isn't enough energy to really make anything explode.  You may get a few sparks is all.  It's not uncommon for me to sit right next to the meter when running.    The half cycle generator is about 600J which can do a bit more damage.   There I short the bus bars before and after running a test.  Risk is much higher (not that 20J is safe my any means).   



 
The following users thanked this post: Faringdon

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
Recent post provides a link to a decent paper on transient testing.   Burst and surge are covered.   
https://www.eevblog.com/forum/testgear/current-transformer-for-esd-testing/msg5587677/#msg5587677

From the paper:
Quote
IMPORTANT SAFETY NOTE: Some of these tests involve electrically hazardous conductors (e.g. mains), and/or
hazardous voltages or energies. These tests can be dangerous, and all appropriate safety precautions must be
taken. If you aren’t sure what safety precautions are needed, ask an expert.

Which is why I would never promote anyone attempt to construct such a generator or use it without having the proper education.   I assume that those with that education could construct and operate such a generator safely.   
 
The following users thanked this post: Faringdon

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7265
  • Country: ca
As a kid I played with ignition coils, neon sign transformers and they educate you very well.
But today? These generators are so expensive, no budget left for training. Electroboom is popular getting zapped. I can't imagine how you can train people on these. If the IQ is low, no amount of yelling will work. I had to work pretty hard to make sure no one got hurt doing the testing. We had a CDN so mains follow-through energy was present which is another level for damage that can happen.

The old Waterfall model for product development, where you test only at the end of a project, certification is a "formality", is your basic disaster. Many companies crash and burn doing this and I delight in watching the engineers facing doing a total redesign, at the last minute. Product development doesn't work that way (Waterfall) unless you have the old guard doing the work.
I would do intermediate stage testing with open boards, no enclosure yet. Basic learning and checking .

Automotive load dump just rented an EmTest and shrapnel was not an issue, it's low voltage in comparison.
 
The following users thanked this post: Faringdon

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
Load dump was a fair amount of energy and with low voltage electronics could do some decent damage.  The amount of testing we perform for new designs that sit on the AC mains is very small compared to what we did while I worked in automotive.       

Quote
The old Waterfall model for product development, where you test only at the end of a project, certification is a "formality", is your basic disaster. Many companies crash and burn doing this and I delight in watching the engineers facing doing a total redesign, at the last minute. Product development doesn't work that way (Waterfall) unless you have the old guard doing the work.

I wouldn't say it's a disaster to have enough products on the market that the majority of your test time is spent running audits.  The amount of time spent testing new designs is small by comparison.  Did the companies you work for not conduct audits?  Did you only test new products being developed?   I worked at a company that followed the one and done model.    Your take on early testing is similar to other places I have worked.   I would say we carry this idea of early testing into other parts of the design, basically designing them to be tested.       

We have a fairly large AC generator that we can switch in for the 50/60Hz.  The surge/burst generator has a CDN as well.  We have damaged stuff that was on the backside of the network.   :-DD   

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 12108
  • Country: us
Quote
Electroboom is popular getting zapped.

Once in a while someone will point out one of his videos to me.  I've seen a few, like that Jacob's ladder tipping over, blab on Walter Lewin and his spark gap transmitter.  I was a big fan of Photonic Induction but I really don't watch a lot of electronic/electrical channels.  Andy's channel was very entertaining as he was playing with things I am never going to see.   


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf