Please login or register.

Login with username, password and session length
gfx gfx
Welcome to ESL DIY!!

This site has been created to provide a forum for DIYers focused primarily on electrostatic loud speaker aka ESL's and the various methodologies associated in achieving highfidelity audio imaging.
gfx gfx
225 Posts in 33 Topics by 25 Members - Latest Member: tuxon September 16, 2014, 05:26:06 PM
gfx* Home | Help | Search | Members | Login | Register | gfx
ESL DIY  |  Transformers, Core, Winding and General information  |  Transformers utilized to power your ESL  |  Testing Your Transformer and Using the Transformer Test Jig

Author Topic: Testing Your Transformer and Using the Transformer Test Jig  (Read 1172 times)

0 Members and 1 Guest are viewing this topic.

Offline geraldfryjr

  • Global Moderator
  • Full Member
  • *****
  • Posts: 84
    • View Profile
In this thread I will be going over a few methods to find the parameters of a transformer using the Transformer Test Jig and with the software tools I presented in this forum here,

I have shown some examples in that thread and here we will be taking a closer look at how to find the parameters that we are looking for.
As well as how they apply when transformers are used as step-up devices in a ESL system.

These same procedures are also used if the transformer is to be used as a step-down device as well for such systems using ribbon drivers.

But this is basically to help explain the mysteries of the almighty step-up transformer for ESL use.

Among the software tools I described in the Transformer Test Jig thread, It is advisable that you have a good standalone signal generator and oscilloscope should you have to encounter any frequency's of about 40Khz or higher.

Also a good frequency counter is almost essentially a must as well for some of these tests to be accurate.

If you have a good sound card that does 24 bit or 16 bit that does do 192Khz sample rate then you may be able to get by quite nicely just using the software tools I presented earlier.

I will be using both in this tutorial as my sound card does run at 192Khz but only reliably in 16bit mode in Visual analyzer and all of the required test functions are built right into the program it self.

I am not going to go over exactly how to operate each of the programs in complete detail and it is up to you to become familiar with them.
But, I will gladly try to answer any questions that you may have to the best of my ability should you have any issues as I did not write these programs.
They are very easy to use and there are plenty of troubleshooting resources on the web where they came from.

What I have done was taken the time to fully test them to make sure that they work for what we are trying to do here.

Our main goal is to is to find out what kind of load our ESL system is presenting to the amplifier.
This is called an Impedance Curve and by using the methods I am about to describe you can plot your curve manually and I will also show how to do this automatically using a few different programs.

Then, I will take our newly found data and apply it to a spice model in a simulation program such as LTspice or Cicuitmaker to check our results.

I will be using Circuitmaker2000 of course as I am most familiar with it and it is extremely easy to use.
This program can still be found on the web if you search hard enough but LTspice is the latest greatest program for electronic circuit simulations and it is free.

LTSpice can be found here,

One last note If you are using a typical digital volt/ohm meter chances are that any voltage measurements that you do at frequency's above about 50hz to 120hz can/will be inaccurate.

I have this problem and it has plagued me to no end, and, is why I have had to revert to using software tools and a good sound card for some of the measurements.
The sound card method is very accurate but it is only good for A.C. voltage measurements and luckily Transformers are A.C. devices.

This won't be an issue if you happen to have a good Meter designed to measure voltages over a very wide frequency range.

Being able measure resistances accurately is a must as well as my meter is off by as much as 10% to 15% or so when switching between ranges and this is very sad.

I sometimes have to use a regulated voltage source such as a LM7805 and measure the current flowing through the resistor and use ohms law to find an exact resistance for my reference resistors.
The ma. current measuring function seems to be fairly accurate on my meter but I still don't have a 100% refference to be completely sure of its errors.

I do use some bridge measurement techniques to assure the accuracy by comparison and good ole' long hand ohms law.
With a calculator of course!!! He,he,he

I can also use VA to do this as well.

I mentioned how I did this in the TTJ thread to measure my home made 1 ohm power resister out of some twisted pair cable to find it is actual value of resistance of about 1.134 ohms or so.

This is the very first meter I have ever owned that I have had this problem with.
It is a Cen-Tec P37772 so if you should ever see this meter and think it is a good deal don't waste your money on it even if it is on sale.

When I got it I had no meter and it has gotten me by with my Variable HV supply project.
As it was during that time that I had discovered its great inaccuracies.
Because of this it has made everything much much more difficult than it really is.

It was however the main reason I had to build the transformer test jig in the first place as I had mentioned in that thread.

More to come soon !!!
Are you ready?

    jer :)

« Last Edit: March 25, 2013, 12:55:34 PM by geraldfryjr »

Offline geraldfryjr

  • Global Moderator
  • Full Member
  • *****
  • Posts: 84
    • View Profile
Re: Testing Your Transformer and Using the Transformer Test Jig
« Reply #1 on: March 29, 2013, 04:22:38 AM »
Let's take a closer look at the transformer that I will be using for for this demonstration.
It started out as a stock Antek AS1206, A 100watt 120v/6v toroid  power transformer that is now slightly modified but still basically stock.
This is the same transformer that I used in the sample plots posted in the Transformer Test Jig thread before it got modified from its stock form as I had received it.

This transformer is/was the shielded version as designated by the AS in its model number.
It was during a high voltage core saturation test I was doing with a rather slow sweep and 40Vrms into the 6V winding and the 120V winding's in series that it failed.
I believe that I may have had the electrostatic shield connected to one side of the HV winding's as well.

Connecting the shield in all of the tests with different configurations only resulted in an increase of primary current.
This shows that it caused an increase of the stray capacitance of the transformer, and, at the highest frequency's this results in an even lower impedance for the amplifier.

Typically this was as much as 120pf to 180pf  without  having any extra capacitance load on the transformer such as an ESL panel.

Antek says in their specs that the transformer is good for at least 3500Vac of insulation resistance as quoted by the data sheet,

" The dielectric leakage
current test is up to 3500Vac in between primary and
secondary coils."

In my test at 40Vrms at around 300Hz or so it finally failed!!!

At the 1:40 transformation ratio I had it wired at, this is only 1600Vac at 300hz and about 2400Vpeak and should not have failed!!!

However it did withstand this much abuse for quite a while before I had hooked up the electrostatic shield, and, as well as a few tests with it connected.

The failure was an arc that went directly to the shield and burned a few turns of the top HV winding closest to the shield causing the turns to connect together and be completely shorted.
Once they were shorted it didn't work at all, and, there was absolutely no output from the transformer as would be expected.

This resulted in about 7 or 8 winding's all together that I found burned and shorted together after I tore it down.

Luckily I was able to clean them off and separate the damaged turns as I recoated each one with some enamel nail polish with teflon.
Then, I used pieces of the 2mil mylar tape that I unwound form the core to further separate the damaged  sections of each turn from one another.
This worked great and I am now able to continue testing with it.

Meanwhile I have not replaced any of the insulation tape that I have unwound from the core and I left off the aluminium foil tape electrostatic shield as well.

I still have the outside magnetic shield that was on it and it consisted of two layers of the core material wrapped around the circumference on the outside of the winding's.
I will add this again later to see if it has any effect on the leakage inductance of the winding's.
Most likely it will be very little if any at all.

For the two 6V winding's there were 26 turns of Bifilar wound 16gauge magnet wire.
I used 13 turns of some common speaker wire to replace it with.
This has a little more DC resistance than the original turns did but the inductance of the winding is not changed.
The self capacitance of the 6V winding is very small to begin with and has no significant effect on the performance of the transformer.

I did mange to break one of the turns while I was trying to repair it and I just bridged it the a jumper wire.
This did not change the number of turns count for that winding and the count stayed as it originally was.

During some lower voltage tests of about 20Vrms to 30Vrms the repaired winding's are holding up well so far.
 I have also tried some short term tests at 40Vrms as well just to be sure everything is working properly.
Everything appears to be in order so now we shall continue.

That is the extent of the minor modifications and here are some pictures showing the transformer in its original state as well as after it was damaged and then repaired.

 jer :) 

ESL DIY  |  Transformers, Core, Winding and General information  |  Transformers utilized to power your ESL  |  Testing Your Transformer and Using the Transformer Test Jig

SMF 2.0.8 | SMF © 2014, Simple Machines Page created in 0.363 seconds with 26 queries.
Helios Multi © Bloc
Powered by MySQL Powered by PHP Valid XHTML 1.0! Valid CSS!