Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Planars & Exotics ESL's, planars, and alternative technologies

Step-up transformer design
Step-up transformer design
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 28th March 2010, 03:29 PM   #151
DamageG is offline DamageG  United States
diyAudio Member
 
Join Date: Jan 2006
Location: So. Cal
Quote:
1. yes, lowering the voltage makes them disappear.
2. lowering the frquency only moves the spikes back down the slope and are consistant with every cycle(i'll take some pictures later).
I think you are seeing crossover distortion. Crossover distortion happens when the current is zero. For a resistive load, this is at zero volts out. For a purely capacitive load, zero current happens at the peaks of the waveform. Sliding down the waveform would indicate the changing phase angle of the load - or something between a pure capacitor and resistor.
  Reply With Quote
Old 28th March 2010, 06:06 PM   #152
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Thank you,damageG, that is along the lines of what i'm thinking.
At the moment I keep getting interupted so I am struggling to keep my train of thought and I will get lots of pics under certain conditions to show you.
This amp is not a big heavy duty one so whatever the reactive compont that is causing this ,it can not supply enough current to over come it.
In my theory antway.
I'm pretty positive though I have a very high leakage inductance with only 10 turns of small gauge speaker wire as the primary wrapped around the four large cores giving me only 10% total coverage tops and ,yes, they are evenly spaced.
Could this be the the reactive component that is giving me problems?
The scope is a 40mhz hitachi V-425 and I have been able to display wave forms as high as 80mhz to 200mhz but not very willingly. jer
  Reply With Quote
Old 28th March 2010, 06:53 PM   #153
bolserst is offline bolserst  United States
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by geraldfryjr View Post
Thank you,damageG, that is along the lines of what i'm thinking.
At the moment I keep getting interupted so I am struggling to keep my train of thought and I will get lots of pics under certain conditions to show you.
This amp is not a big heavy duty one so whatever the reactive compont that is causing this ,it can not supply enough current to over come it.
In my theory antway.
I'm pretty positive though I have a very high leakage inductance with only 10 turns of small gauge speaker wire as the primary wrapped around the four large cores giving me only 10% total coverage tops and ,yes, they are evenly spaced.
Could this be the the reactive component that is giving me problems?
The scope is a 40mhz hitachi V-425 and I have been able to display wave forms as high as 80mhz to 200mhz but not very willingly. jer
Jerry,

Leakage inductance itself is not a problem, but in combination with the capacitive load from the winding capacitance, it forms a series resonant circuit that will drop the impedance to VERY low values(easily 0.5 ohm or less) at high frequencies. I measured the winding capacitance for a few larger toroidal power transformers I have laying around at about 1500pF. That might explain why you aren't noticing much difference when you add your 80pF ESL load. Have you measured the impedance vs frequency for your transformer setup yet to see what the minimum impedance is and at what frequency?

As Alex mentioned, measuring the voltage across a current sensing resistor put in series with the ground wire from you amplifier will help you figure out what it is going on. You mention your amplifier is not particularly heavy duty, so the voltage spikes you see might be it's way of protesting against the low impedance load. Try a larger value of series resistor like 4 ohm to see if this helps out. This will roll off the top octave response a bit, but will give you an idea if it is the low impedance load(combined with high phase angle) that is causing a problem for your amplifier.

Concerning one of your other observations: You said...
"I also then split my primary into two 5 turn sections.
Using one 5 turn winding I adusted the setup for the pulse and by adding the other 5 turn winding in parallel with the first one. The pulse got smaller."

Once you added the 2nd set of 5 turns in parallel and reduced the leakage inductance, the resonant frequency for minimum impedance would have moved up in frequency. With the 2nd set of 5 turns added, try increasing the frequency to see if the spike level increases.

Keep at it, systematically testing stuff.
You'll get it figured out, and learn a bunch in the process.
  Reply With Quote
Old 28th March 2010, 08:16 PM   #154
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
Thanks steve, i am trying to keep it as systematic as i can.
I don't have a capacitance meter yet and sure i wish i had one right now!
yes,with both primary's connected at 30khz i adjusted for a very small spike.
it disappears as i move up to 40khz and then reappears when I disconnect one of the windings.
also i have tried larger value resistors but that greatly reduces my output to the speaker. jer
  Reply With Quote
Old 28th March 2010, 09:15 PM   #155
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
As I was testing above 20khz was fighting to achieve a flat response with my mixer even though it will pass 100khz with no problem.
so i decided to eliminate it from the signal path.
the amp doesn't go into shutdown any more and i am able to push it into clipping with no problem.
the pulse is still there but not as bad now without the mixer .
the pulse resonance seems to be around 50 khz,just as the amp hits it's rail it goes into parasatic oscillation and starts to draw alot of current and then shuts down.
I'm going to go back over everything and try the primary thing and what ever else I think of as I am determined to solve this problem.so keep the suggestions coming as I don't quit easily .jer
  Reply With Quote
Old 28th March 2010, 09:42 PM   #156
bolserst is offline bolserst  United States
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by geraldfryjr View Post
also i have tried larger value resistors but that greatly reduces my output to the speaker. jer
But with a larger resistor, like 4 ohm, does the spike still appear?

If you haven't already done this, why don't you get an estimate of your impedance vs frequency at the top of the audio band. You can do this be measuring the voltage across the series resistor and the primary at various frequencies. I usually use a 10 ohm resistor, but anything from 4 ohm to 30 ohm will work fine. start at 1kHz and work your way upward. Measure the voltage across the series resistor, and then across the transformer primary. The transformer impedance at that frequency can then be estimated by:

Z = R x VT / VR

R = value of series resistor
VT = voltage measured across transformer primary
VR = voltage measured across series resistor

As you work your way upward from 1kHz, you will probably see your impedance steadily falling till you reach a minimum (at resonance between leakage inductance and winding capacitance) and then rising again after than.

With a 10 turn primary the toroidal power transformer I have reached a minimum impedance of 0.38 ohm at 16kHz.
That's a tough load for any amplifier.

Last edited by bolserst; 28th March 2010 at 09:53 PM.
  Reply With Quote
Old 29th March 2010, 12:14 AM   #157
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
steve,what voltage level should I use?
At a set frequency the voltage ratio changes when I change the voltage level.as in a earlier test when i measured a turn ratio of 160:1 at a low voltage level and a turns ratio of 130:1 at a high voltage level.
strange? jer
  Reply With Quote
Old 29th March 2010, 02:38 AM   #158
bolserst is offline bolserst  United States
diyAudio Member
 
bolserst's Avatar
 
Join Date: Jun 2005
Location: Fort Worth, Texas
Quote:
Originally Posted by geraldfryjr View Post
steve,what voltage level should I use?
At a set frequency the voltage ratio changes when I change the voltage level.as in a earlier test when i measured a turn ratio of 160:1 at a low voltage level and a turns ratio of 130:1 at a high voltage level.
strange? jer
I would use 1 to 2 Vrms for impedance measurements.
The calculation involves the ratio of the voltages, so that actual value doesn't matter as long as you don't change the input voltage between measuring the voltage across the series resistor and the transformer primary.

I'm not sure the cause of your turns ratio discrepancies without more information.

1) What frequency were you doing the measurements? To avoid contaminating your measurements with the resonant interaction of leakage inductance and winding capacitance I'd recommend using a low frequency of 100 to 500 Hz. If you use a frequency near the resonant frequency mentioned above and are not using a series damping resistor you can measure an "effective" turns ratio of 2 to 3 times that at low frequencies. This is the cause of the SPL peak seen at the top end of the audio band on most ESLs. You need to choose a damping resistor to tailor the response to your liking. You will see increasing values of "effective" turns ratio as you approach the resonance. At the frequency of maximum peaking, you will measure the minimum impedance, and the maximum current drawn from your amplifier. Just below this frequency is where many amplifiers have stability issues as the impedance is still quite low, and the phase can be -60 deg or more.

2) Were you using a series resistor? If yes: Were you measuring the input voltage for the ratio calculation before or after the series resistor. You should measure after the resistor, directly across the primary winding.

3) Were you watching the waveform on the oscilloscope to make sure that distortion or clipping would not affect the voltage reading of your meter?


As a side note, one thing that does change with magnitude of input voltage is the impedance at low frequencies. This is because the inductance depends on the permiability of the core which is a function of flux density. In general, inductance and impedance will slowly increase with increasing input voltage until saturation is approached. Then further increasing the input voltage you bring about rather abrupt decrease in inductance and impedance as the core saturates.

I mention the above because many ESLs, including the Audiostatic ES100 use a series capacitor to create a 2nd order HP filter with the primary inductance of the transformer. As you can imagine, without adding some components(resistors or inductors) to swamp the variable inductance of the step-up transformers you will have a crossover that changes response with input level.

Calvin has some excellent posts on how to properly create a HP network for an ESL.
http://www.diyaudio.com/forums/plana...tml#post727647

Last edited by bolserst; 29th March 2010 at 02:50 AM.
  Reply With Quote
Old 29th March 2010, 02:43 AM   #159
alexberg is offline alexberg
diyAudio Member
 
alexberg's Avatar
 
Join Date: Oct 2009
Quote:
Originally Posted by geraldfryjr View Post
As I was testing above 20khz was fighting to achieve a flat response with my mixer even though it will pass 100khz with no problem.
so i decided to eliminate it from the signal path.
the amp doesn't go into shutdown any more and i am able to push it into clipping with no problem.
the pulse is still there but not as bad now without the mixer .
the pulse resonance seems to be around 50 khz,just as the amp hits it's rail it goes into parasatic oscillation and starts to draw alot of current and then shuts down.
I'm going to go back over everything and try the primary thing and what ever else I think of as I am determined to solve this problem.so keep the suggestions coming as I don't quit easily .jer
Two cents,
spike seems to be voltage dependent - so I assume the worst case - partial or barrier discharge due to insulation not designed for high voltages. No answer on what voltage it begins... If it moves to the beginning of the cycle as frequency goes down - you need a certain volt-second product to accumulate the charge on insulation layer then it breaks down.
As others suggested it well might be crossover due to the inductive impedance, again you need certain volt second product to recoup the energy in the load to the power supply.
The latter, however, shall be present at any frequency and voltage level, apmlitude should vary.
Can you load your transformer at the secondary? It looks hard to do, but it is not - put about 20 2W resistors in series with total resistance around 100k to 300k and burden the secondary. Impedance, reflected to the primary should become mode dumped out or resistor like and see how it goes.
Once again, can you swap transformers?
Alex
  Reply With Quote
Old 29th March 2010, 07:39 AM   #160
geraldfryjr is offline geraldfryjr  United States
diyAudio Member
 
Join Date: Feb 2008
Location: Jackson,michigan
sorry, i didn't get back to you as was trying every possible combination with the windings I could and had a few issues along the way.
the pulse could be crossover switching and causing some rf.
I fried a cuple of meters and two 470 ohm resistors that were in series and a 120v winding on each end forming the center tap in the middle and nothing connect to the other end of the windings.
first the resistors and then the meters then I knew it was rf!
I did start with a new transformer and it does the exact same thing.
then I got two of them and made sure they were identical and not mismatched.
Taped them up and wound 10 turns on them.
here's what i found:
transformer A ,120v (1),120v(2)
transformer B ,120v (1),120v(2)
center tap point ,ct
A(1)+CT+A(2) problem
B(1)+CT+B(2) problem
A(1OR2)+CT+B(1 OR2) no problem at 20 khz
A(1)+B(1)+ct+A(2)+B(2) problem swap (1) for (2) same
only when two windings from the same transformer are connected together does the problem occur.
with all of the windings left open resonance is 50khz.
Then I tried connecting all the windings on each in series with each other this loads the amp down severly 20khz and up with dullpeak at 32.5 khz
Try to connect for a center tap and it draws and maintains an rf type arc with 20v p-p input with nothing connected to the other ends.
it is left unconnected at this point.
the pulse starts to come in at 3khz at 80v p-p and does not change position until a resonate peak of 22.5khz at which point it is positioned at the peak of the sine wave.


tommorrow i will plot the impedence curves to look at. jer
  Reply With Quote

Reply


Step-up transformer designHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Step Down Transformer Ropie Power Supplies 3 15th October 2006 09:42 AM
Can I use a step-up transformer as step-up? cviller Parts 5 6th September 2006 06:32 PM
help design SE preamp with mosfet buffer and step-up transformer. AudioGeek Solid State 0 19th January 2005 03:02 AM
Help Newbie Design X-over Step-by-Step Kongen Multi-Way 3 21st November 2004 02:32 AM
Using one transformer to step down another. Shoog Solid State 9 28th August 2003 10:00 PM


New To Site? Need Help?

All times are GMT. The time now is 02:55 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki