LA2A with Edcor Transformers - PROBLEMS!!!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

khstudio

Well-known member
Joined
Dec 27, 2005
Messages
2,116
Location
New Jersey, USA
I just used the Edcor XSM Transformers in my LA2A.
Everything in the circuit is CORRECT!
I've checked & rechecked several times.
NOTE: I'm using the DRIP v2 PCB's for this build.

ALSO, notice the ROLL-OFF starting under 1k.
It's definitely NOT the adjustable cap, C4!

C4 is set to 150pf for ALL the Pics Below (except my original)

INPUT = XSM 10k:10k
OUTPUT = XSM 10k:600

When I was sweeping it with RMAA I got VERY strange LOW END behavior. :?
Un-Loaded_output_%2B_150pf_C4.jpg


My first 2 LA2A's used JENSEN Tranny's... same RATIO's for I/O & they look & work fine!
My_Original_LA2A.jpg


I was able to tame it down to normal by putting an 11k resistor on the OT's Primary & a 600 ohm across the secondary.
Loaded_output_%2B_150pf_C4.jpg


Here it is LOADED with JUST the 600 ohm on the secondary:

Loaded_600_only_%2B_150pf_C4.jpg


this is not normal...
I've read a lot of you guys use these tranny's but never read about this, loading or any problems like this.

And more important, what could be causing the HIGHs to start rolling off so early?

PLEASE HELP! :sad:
 
did you try the circuit without transformers? it can be done, just run unbalanced in and out. dont blame the transformers till youve eliminated the rest of the circuit, the test setup etc

i have used the same xsm models in a couple of circuits and they did not cause the kind of problems youre having, frequency response was flat.
 
[quote author="[silent:arts]"][quote author="khstudio"]Added pics & more detailed info to 1st post.[/quote]
you told us you don't hear it - where is the problem ?[/quote]

:? look at the pictures. What kind of question is that?
Look again... I took that sentence out of my first post.

After more listening & comparing to my first build you can sense somethings a little off.

I didn't start this thread because I'm bored... I'm having a problem & want to fix it & share my findings.
 
[quote author="solder_city"]did you try the circuit without transformers? it can be done, just run unbalanced in and out. dont blame the transformers till youve eliminated the rest of the circuit, the test setup etc

i have used the same xsm models in a couple of circuits and they did not cause the kind of problems youre having, frequency response was flat.[/quote]

Great :thumb: I was thinking about doing that.
How should I go about hooking it up to my BALANCED sound cards I/O for testing?
 
soundcard output, use just pins 2 and 1 on the xlr, leave pin 3 float. this will work if the output is ground referenced, which is 99.9% likely (doubt they transformer balanced the output!)

soundcard input, use pin 2 for hot, use 1+3 (connected together) for cold.

warning: the output level of the la-2 will be very hot because of no stepdown loss from the transformer. and be very sure you take it from the 'output' side of the coupling cap! :shock:
 
warning: the output level of the la-2 will be very hot because of no stepdown loss from the transformer. and be very sure you take it from the 'output' side of the coupling cap!

I was JUST going to ask about that. :thumb:

It's a LYNX ONE AD/DA Card.
I've never seen this Low end or High roll off problem with any of my other gear I've tested.

One thing to note, I'm not using the GRID LOAD resistors on the 12bh7...
Could that cause ANY of these problems... Oscillation :?:

I'm also using this DRIP LA2A PCB & have check EVERYTHING! :?

Thanks for helping, I just want to get to the bottom of it.
 
[quote author="khstudio"] :? look at the pictures. What kind of question is that?
Look again... I took that sentence out of my first post.[/quote]
why - did you hear it after looking at the pictures ?
I don't read all posts from the beginning to see if one is updated ...

loaded with 600R you have +3dB at 4Hz, starting at 10Hz :shock:
this is of course a major problem. :wink:
of course there will be a difference to the Jensens, or any other Transformer. or any other build unit (+/- XX%).
otherwise, there shouldn't be a price differnce ?
I have the same Edcors handy at the moment, will measure them tomorrow :thumb:
 
One more thing,

Do you guys that use these EDCOR's use the CENTER TAP?
Should I be on the LA2A?

I'm not right now... just using the + & - on the primary & secondary.
 
[quote author="[silent:arts]"][quote author="khstudio"] :? look at the pictures. What kind of question is that?
Look again... I took that sentence out of my first post.[/quote]
why - did you hear it after looking at the pictures ?
I don't read all posts from the beginning to see if one is updated ...

loaded with 600R you have +3dB at 4Hz, starting at 10Hz :shock:
this is of course a major problem. :wink:
of course there will be a difference to the Jensens, or any other Transformer. or any other build unit (+/- XX%).
otherwise, there shouldn't be a price differnce ?
I have the same Edcors handy at the moment, will measure them tomorrow :thumb:[/quote]

I didn't mean to be smart & could use all the help I can get.
I brought this up because I'm not so sure everyone actually measures their gear. But with the LA2A, you have to, to set the adjustable caps & the high roll off point... mine is just rolling off WAY earlier than the cap! Not to mention the LOW issue.

And your right, I thought I'd save a few bucks & use these EDCOR's but before I give up I'd like to see if they're the problem.
... getting ready to run it UN-BAL to see.
 
I also forgot to mention I spoke with Brian @ EDCOR today & told him what I found... He said it was probably a good idea to load the OT on both side like I did.
He explained why but it was over my head :oops:
 
since I will use the same transformers in my Analag Opto prototype I have them here. will check them tomorrow. :thumb:
(but I still think you shouldn't compare it with your jensens)
 
Adding the GRID resistors on the 12bh7 output stage DID NOT HELP or change anything. :sad: :guinness: :guinness: :guinness:

Time to try Un-Balanced/ no trannys.

I'll try the OT first.
 
[quote author="solder_city"]did you try the circuit without transformers? it can be done, just run unbalanced in and out. dont blame the transformers till youve eliminated the rest of the circuit, the test setup etc

i have used the same xsm models in a couple of circuits and they did not cause the kind of problems youre having, frequency response was flat.[/quote]

How did you load them & what circuit was it?

& may I ask, how would you wire them & set them up on an LA2A circuit?
 
[quote author="khstudio"][quote author="[silent:arts]"][quote author="khstudio"]Added pics & more detailed info to 1st post.[/quote]
you told us you don't hear it - where is the problem ?[/quote]

:? look at the pictures. What kind of question is that?
Look again... I took that sentence out of my first post.

After more listening & comparing to my first build you can sense somethings a little off.

I didn't start this thread because I'm bored... I'm having a problem & want to fix it & share my findings.[/quote]

I agree it doesn't seem right, and I also agree with 'where is the problem?'. Now, the Edcors should do better than that because I chart them better than that. But, I can show you plenty of killer sounding antique limiter charts that would make you think you'd showed us a ruler flat response in comparison. Really. So, if you can't ever find that something's actually wrong, you get to choose whether or not you like the color option, or if you need to change the iron. Better yet, swap some of your fancy Jensen iron into this one and see what happens. That might rule it all out. And it will do so faster than we can all sit here and conjecture.

Other idea; reverse leads on the output primary, then on the input secondary and check each for better or worse response. figure out polarity later when looks best.
 
UPDATE :!: :!: :!:

This is the output ONLY UN-BALANCED!
UN-BALANCED_OUT.jpg



This is BOTH INPUT & OUTPUT UN-BALANCED!

It looks PERFECT. :shock: :green: :?

UN-BALANCED_IN___OUT.jpg


So... I now would like to fix it so it will work or at least understand why?
I set the inputs & output up like the JENSEN MOD'S & used their values for loading.
The schematic is HERE:
http://www.jensen-transformers.com/as/as091.pdf

The Jensen input is 10k:10k & they recommend an 11k across the secondary... so this is what I used with my EDCOR 10k:10k inputs.

As you can see in the schematic (for the output) Jensen recommends loading the Primary with 11k IF the Secondary is 600 ohms.
Well, from my Pics in my first post you can see that I needed to do both to tame the 10Hz to DC rise.

So it IS the Trannys causing problem. :sad:

ANY advise is welcome & needed.

Thanks.
 
you can adjust those resistors until flattest also.

but still flip polarity like I said earlier and look at that.

are the transformers totally disconnected in each picture?

try leaving them connected and run signal as if not, meaning signal into input sec and out from output pri. see what effect the iron has on the chart like that, where it's a loading effect rather than having signal pass through.
 
are the transformers totally disconnected in each picture?

Pic 1 is just the input fully connected & output totally removed.

Pic 2 is BOTH totally removed.

So it looks like 99% of the BS is from the input.

you can adjust those resistors until flattest also.
Which ones?
There's only 1 left (on the input side) after removing the OT... the loading for the OT was directly on the OT.

try leaving them connected and run signal as if not, meaning signal into input sec and out from output pri. see what effect the iron has on the chart like that, where it's a loading effect rather than having signal pass through.

I thought about that too... I'll give it a try on the input (since that's where the issues are)
 
Back
Top