Sanity Check - Input Transformer Overshoot - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 11th December 2012, 02:40 PM   #1
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Default Sanity Check - Input Transformer Overshoot

Hello all, just checking my logic.

I have built a 26 preamp using an LL1676 input transformer. I was able to nicely damp the transformer ringing on square waves using a 2nF+850ohm RC network, but I still had about 25% overshoot on the very front of the waveform when feeding the transformer from my function generator (Zout=50ohms).

Adding a 950ohm resistor in series with the + input to the transformer completely killed the overshoot and resulted in a near-perfect square wave. The total effective output impedance of my function generator was now 50ohm + 950ohm = 1000ohm, to achieve a perfect square wave. Of course, all my sources will have different output impedances, for example my CD player has 200ohms and my phono stage has 400ohms.

I'm thinking that the best way to handle this would be adding extra resistance to the outputs of my sources so that their total output impedance always = 1000ohms (so for the CD player that would be 800 ohms+the normal 200ohms, for the phono stage that would be 600ohms plus the normal 400 ohms) instead of adding resistance to the input transformer. In this way, no matter the source I'll always have the correct impedance for a perfect square wave.

Correct??
  Reply With Quote
Old 11th December 2012, 02:55 PM   #2
diyAudio Member
 
Join Date: Dec 2005
Location: Kuala Lumpur
Most sources don't generate square waves or anything much above 20 kHz, so the overshoot is unimportant
  Reply With Quote
Old 11th December 2012, 03:33 PM   #3
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
The LL1676 datasheet says it is for 600ohm source and 10K load, when wired for 1:4 mode. In this case it would present a load impedance of 600ohms, so not for normal domestic stereo line stage use but receiving from a balanced line. If you wired it as 1:1, then my guess is that it would match 2.5k to 2.5k - again quite unsuitable for normal line stage use.

So how are you using it? My guess is that you are seeing the HF resonance at 70kHz.
  Reply With Quote
Old 11th December 2012, 03:43 PM   #4
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by DF96 View Post
The LL1676 datasheet says it is for 600ohm source and 10K load, when wired for 1:4 mode. In this case it would present a load impedance of 600ohms, so not for normal domestic stereo line stage use but receiving from a balanced line. If you wired it as 1:1, then my guess is that it would match 2.5k to 2.5k - again quite unsuitable for normal line stage use.

So how are you using it? My guess is that you are seeing the HF resonance at 70kHz.
It's wired backwards from the datasheet at 2:1, in a 2||2:1||1 setup. In this way I get suitable input voltage from a line source, I can invert the phase to compensate for the inversion of the 26 tube, and I get galvanic isolation to prevent ground loops (it's really a quiet preamp!). Input impedance measures 74kohms at 100Hz, and frequency response is flat below the limit of my spectrum analyzer (10Hz).

I'm happy with the sound as is but would like to get rid of the overshoot and have succeeded with the added input resistance as described above, so I'm just checking my logic that adding resistance to each source may be the best way to do that.

Last edited by Magz; 11th December 2012 at 03:46 PM.
  Reply With Quote
Old 11th December 2012, 03:58 PM   #5
pooge is offline pooge  United States
diyAudio Member
 
Join Date: May 2003
Location: Northern Va.
Quote:
Originally Posted by Magz View Post
I'm thinking that the best way to handle this would be adding extra resistance to the outputs of my sources so that their total output impedance always = 1000ohms (so for the CD player that would be 800 ohms+the normal 200ohms, for the phono stage that would be 600ohms plus the normal 400 ohms) instead of adding resistance to the input transformer. In this way, no matter the source I'll always have the correct impedance for a perfect square wave.
IF you went this route, it would be better to insert the resistors at the inputs of your preamp, rather than the outputs of your sources, so they won't negatively interact with the capacitance your interconnects.
  Reply With Quote
Old 11th December 2012, 04:00 PM   #6
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
I think your problem may be that you are just using it in too high an impedance situation. As wired, it would want to match 2.5k to 600ohms - far too low for normal line stage use. As the HF resonance is well away from the audio band you can probably run at a bit higher impedance, say 5k to 1.2k or even 10k to 2.4k. This is still quite low for a line stage. Your 74k input presumably means you have something like 18k load on the output. Make this much smaller.

Adding a snubber will make the input impedance become rather reactive.

You really need either a higher impedance transformer, or no transformer at all.
  Reply With Quote
Old 11th December 2012, 04:08 PM   #7
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by pooge View Post
IF you went this route, it would be better to insert the resistors at the inputs of your preamp, rather than the outputs of your sources, so they won't negatively interact with the capacitance your interconnects.
Ah, I thought there might be a reason not to put it at the sources, but couldn't put my finger on it. Thanks, that answers my question...I'll put 800ohms at the input and be done with it. Thanks!
  Reply With Quote
Old 11th December 2012, 04:57 PM   #8
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Quote:
Originally Posted by DF96 View Post
I think your problem may be that you are just using it in too high an impedance situation. As wired, it would want to match 2.5k to 600ohms - far too low for normal line stage use. As the HF resonance is well away from the audio band you can probably run at a bit higher impedance, say 5k to 1.2k or even 10k to 2.4k. This is still quite low for a line stage. Your 74k input presumably means you have something like 18k load on the output. Make this much smaller.

Adding a snubber will make the input impedance become rather reactive.

You really need either a higher impedance transformer, or no transformer at all.
You're right. I need a purely resistive load of about 1k for a smooth response, but the input impedance is way too low then. Using an RC with a corner frequency of 94k takes care of the ringing without affecting the audio range. Adding the series resistance fixes the remaining overshoot to give a nice square wave at 2kHz.

Maybe a little unorthodox, but it sounds good.
  Reply With Quote
Old 15th December 2012, 05:34 PM   #9
Magz is offline Magz  United States
diyAudio Member
 
Join Date: Oct 2009
Location: New Jersey, USA
Here's how a 10KHz square wave looks on the LL1676 wired 2||2:1||1 with no load, with an RC of 2nF and 850 ohms on the secondary, and with the RC plus 950 ohms in series with the primary.

I think the third pic looks pretty good! Input impedance measured with LCR meter at 83Kohms at 100Hz.
  Reply With Quote
Old 15th December 2012, 07:02 PM   #10
diyAudio Member
 
HollowState's Avatar
 
Join Date: Aug 2006
Location: Taxland, New Jersey
Why do people insist on using a square wave frequency of 10K? With the harmonics involved in making up a square wave, you are demanding a transformer have a usable flat bandpass of 100KHz or more. This is beyond the capabilities of a Lundahl LL1676. A 2KHz square wave is the proper frequency that will equate out to 20KHz.

If your signal generator has an output impedance of 50 ohms, then you should use a 50 ohm characeristic cable and terminate that cable, where it enters the device under test, in 50 ohms. Anything else higher will invite aberrations like overshoot. Any padding to compensate the waveform is unreliable because it may mask an underlying condition.
__________________
"It is hard to imagine a more stupid or more dangerous way of making decisions than by putting those decisions in the hands of people who pay no price for being wrong." ~Thomas Sowell
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
2-Way Sanity Check wrenchone Multi-Way 8 9th December 2011 07:54 PM
Capacitor Sanity Check treagle Tubes / Valves 6 19th March 2011 10:06 PM
Sanity Check ..... please Lostcause Digital Source 0 16th October 2008 06:19 PM
Sanity check on an external transformer box. InSides Analog Line Level 1 9th January 2008 03:33 PM
Sanity check planet10 Multi-Way 14 7th March 2004 10:15 PM


New To Site? Need Help?

All times are GMT. The time now is 06:44 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2