1272 sounds thin

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

datape

Well-known member
Joined
Feb 13, 2006
Messages
63
Location
UK
I'm trying to wire up a BA183 using the diagram from the JLM site.

http://www.jlmaudio.com/JLM1272simplemod.pdf

Except I have used the larger 4700uF capacitors on the N and the M pins

The output is the recent custom wound Carnhill job.

But I am using the unheard of Amethyst Designs input transformer.

It is rated as a direct repacement for the 10468, with the same values for primary and secondary windings.

In theory this should work, but I have nothing to compare it with to see if it is suitable.

The problem is the audio is very weak and thin. Am I not loading the transformer properly using the diagram?

The transformer only has 8 wires coming from it.

I have guessed that the wires measuring the lowest resistance with a digital meter are the primaries. And the higher resistance are the secondaries. And simply connected them in series on each side.

Have I missed something or is it the transformer?

It also hums quite nastyily, but I figured i'd get the audio sounding right before I worried about that.

Please someone help if they can.
 
Test the transformer and the preamp seperately, to isolate the issue. Do you have a scope or a test application like Rightmark?

Run a signal direct into the 5k pot, and see if it comes out intact. You can put together a simple pad setup to run a line level signal into the transformer, terminate it properly and test it's frequency response. NewYorkDave provided a diagram for such a setup. Lemme see if I can find that thread....

I'm just getting started on my 1272, stuffing the 283 board, using groupbuy iron as well.
 
[quote author="datape"]Except I have used the larger 4700uF capacitors on the N and the M pins[/quote]

Never heard of using 4700uF on the N pin (only on the M pin). N usually gets a 470 to 680uF.

See Step2: http://glideonfade.com/aswiring.htm
 
Okay I downloaded the Rightmark software.

I guess I need to use some kind of resistor network to test the frequency of the transformer without the circuit.

Would I need this on both sides of the transformer?

Also can I patch a line from my soundcard straight into the circuit at the 5k pot to test it?

RE the caps I changed the 4700uF on the N to a 470uF but it didnt make a difference.
 
[quote author="datape"]Okay I downloaded the Rightmark software.

I guess I need to use some kind of resistor network to test the frequency of the transformer without the circuit.

Would I need this on both sides of the transformer?[/quote]

Yeah, a pad on the input side. Try a U-pad that reflects 150 ohms impedance into the transformer. I think a pair of 1k resistors in line with the input signal and a 220 across the primary should work, your basic 20db pad. Could be a pair of 680s and a 160.

The secondary needs a load resistor approximating the impedance it will see in circuit. Try a 5k resistor in place of the 5k pot. Keep the 180pF before that.

[quote author="datape"]Also can I patch a line from my soundcard straight into the circuit at the 5k pot to test it?[/quote]

Yeah, I don't see why not. with the pot you can simply turn down as much as needed.


[quote author="datape"]RE the caps I changed the 4700uF on the N to a 470uF but it didnt make a difference.[/quote]

Yeah, I didn't think that would fix your bass response problem. I'm actually unclear on why these cap values are so critical. People seem to have strong opinions on what values sound best. I would have expected there to simply be a minimum to make enough current available to the circuit, and more would be fine. If that's the case, your larger value wouldn't hurt anything.
 
I would definitely try injecting a signal at the 5k attenuator pot (the one hanging off the secondary of the input transformer). If that fixes the problem then you know the problem is at the input transformer. I think you may have gotten the polarity backwards on one or more of the input transformer windings. Does the input transformer have split primary and secondary like the 10468? If so, you'll need to first try swapping pins 2 & 4 (referring to the JLM diagram). If it's still bad, then leave the primary swap and swap the ends of one of the secondary windings, if it's still wrong leave the secondary swap and unswap the primary. That should cover all possible polarity cases...

A P
 
Ok, the circuit is clean. It passes line level audio directly into the 5k pot.

As soon as I connect the transformer it starts humming.

These transformers only have the wires for the primaries and secondaries, no connection for an earth and they are inmouled plastic cases. I'm wondering if they are not so ideal with the humming let alone with the other issues.

I tired swapping polarity on the input and that competely lost the signal so I guess it wasnt that.
 
If you tried swapping polarities of the input transformer and lost the signal entirely then you either have a bad transformer, one ill suited for the purpose or haven't identified the primaries and secondaries correctly...a transformer will work in either polarity...

Cheers,
Ray
 
i'll go back and check the polarity of the transformer, but I swapped pins 2 and 4 as suggested and lost all signal. I'll check my wiring again.

but do you think is it worth messing around with these given the hum? The original 10468 has some sort of earthing system doesnt it.

I may contact ameythest designs and see why they left the earthing off
 
The only way the signal would completely cancel is if you ran dual primaries in parallel out of phase. Lemme try a CJ-style paint image to illustrate:

dualpri.PNG


Maybe the Amethyst transformer is internally connected to its case, so you could just connect the case to chassis?
 
If you could answer the questions I asked previously, it would help figure out the problem. Also, you only did the first step of the three step experiment I outlined.

How many pins does the transformer have?

Can you measure the resistance between each pin and all the others and build a table like this (for a 6 pin transformer--add rows/columns as necessary for your device)?

Code:
    1    2    3    4    5    6
1   0   100  open open open open
2        0   open open open open
3             0    99  open open
4                  0   open open
5                       0    27
6                            0

Then it should be obvious which pins connect to which windings and if there is a short or open in any of the windings. The hum may not be a problem if you end up building the pre in a metal enclosure. Also, the transformer may have humbucking windings and you've got them mis-wired.

I'm glad to try to help, but you need to give us more info than you have up to this point.

A P
 
aplogies AP for not answering. I keep getting pulled away and coming back to it.

anyway the transformers have 8 wires coming from them, the readings in ohms resistance are as follows

RED + YELLOW 23.4
GREEN+ORANGE 23.4
WHITE+PURPLE 146.4
GREY+BLACK 145.8

This approx the same reading as CJs post

http://www.groupdiy.com/index.php?topic=13601&highlight=10468

I originally connected YELLOW and GREEN together and connected RED and ORANGE as the XLR input

I then connected PURPLE and GREY together and connected WHITE to the 5k pot input and BLACK to the 0v connection on the pot. I also had a 180pf capacitor across the BLACK and WHITE wires.

I will try and swap the polarity of the all the wires. HOWEVER I get hums as soon as I connect the transformer wire to the 5k pot. Perhaps I have them miswired.

The transformers are in a moulded black plastic case so there is no way an internal earth could be connected to the case
 
[quote author="skipwave"]The only way the signal would completely cancel is if you ran dual primaries in parallel out of phase. Lemme try a CJ-style paint image to illustrate:

dualpri.PNG


Maybe the Amethyst transformer is internally connected to its case, so you could just connect the case to chassis?[/quote]

Right, but he's getting something out. Skip--you left out one possibility which is what I was gueesing might be going on here. Referring to your series diagram--what if datape has 2 going to +, 3 going to -, and 4-5 bridged. Now he's got a series connection with the windings out of phase.

Datape: Did you check the resistance between windings to make sure there's not a fault in the transformer? Please fill out the full table. Don't assume anything yet!

A P
 
sorry I did check all the measurements and I have ten of these transformers, i've tried half and they all behave the same.

I just tried something else though

I connected only half the transformer ie RED and YELLOW to the XLR and GREY and BLACK to the 5k pot.

Well it doesnt hum, but I'm not getting the right ratio am I?
 
The ratio is the same, but the DCRs are halved and the inductance is much lower (sqrt?). Does it now pass audio?

Going back to my original suggestion, you need to make sure you've got both the primary and secondary windings connected in series with the correct polarity or it won't work. You have only tried 2 of the four possibilities so far. I think this is the problem.

A P
 
Finally got it. It was something to do with the polarity of the windings

I connected YELLOW and ORANGE together and the GREEN and RED to the XLR. Then BLACK and PURPLE together and GREY and WHITE to the 5k pot.

Passes audio pefectly.

Thanks for alll the help and suggestions everyone.
 

Latest posts

Back
Top