Making a BD139 voltage follower buffer, need a little help

[~n00beR]

Some kind soul allowed me to look under the cover of there canamp.

The schematics from around the web appear and my own (from pictures) are wrong.

My schematic shows R4 at the emmiter of the transistor, as amb and majkel have suggested it should be between the opamp and transistor, this is also the correct configuration in the canamp.

The amp i managed to get a look at has been modified and has had the input capacitors removed, with no measurable problems, as well as having a regulated PSU and upgraded caps and rectifer diodes.
It sounds very nice indeed.

However i couldn't figure out what the capacitor that sat at the top of the opamp is for?!?

Anyway glad to see your amp is working

 

[majkel]

5 seconds with no harm to the finger means the transistor will survive.

 

[FallenAngel]

Thanks for the advice, but being overanxious as I am, I already cut this one to size. Pics soon

 

[FallenAngel]

Hey guys, a question regarding changing the power supply in this amp. If I was to make this amp with a virtual ground off a single secondary 24V AC transformer, which virtual ground scheme would you recommend? Since this amp has a huge output current would the easiest thing be to simply add another opamp + BD139 to drive the ground and would it need to pull double the current?

 

[amb]

Quote:

Originally Posted by FallenAngel /img/forum/go_quote.gif Since this amp has a huge output current would the easiest thing be to simply add another opamp + BD139 to drive the ground

Yes. Better yet, do it M³/Mini³/PPA/Pimeta style. Use a TLE2426 as rail splitter for the nil-current signal ground, and use your 3rd channel to drive the output ground return from the headphone.

Quote:

and would it need to pull double the current?

The ground channel will source/sink the combined return current from the left and right channels, and if such signal is in-phase mono, then it would indeed be double. But the BD139 output stage has plenty of current capability so it doesn't really matter.

 

[FallenAngel]

Thanks amb, I'll be making a 3-channel of this soon and will try it using the same ground scheme as those.

I just noticed something today when I was changing the gain on this amp. I messed up with one connection and locked the gain to about 2.

This is the schematic as it should be:

This is what I actually connected:

Now I'm not sure what this actually did and it does sound great, but the gain is too low and I wanted to raise it. When I "fixed" the amp by making it exactly by the schematic, I got a very nasty buzzing noise in the headphones. Still no offset, just this irritating buzzing noise. Not really a hum, a definite buzz.

Any ideas on what to do with it?

 

[amb]

The wiring mistake yielded a unity-gain amp. Not gain of 2.

As for the buzzing, is your amp enclosed in a metal case and properly grounded? How far away is the power supply and its transformer from the amp? By the way, the "fixed" version has a gain of 11. If you have low-Z or sensitive headphones it might be too much gain, and you'd hear noise that might otherwise go undetected in a high-Z phone.

 

[FallenAngel]

Thanks amb, I guess the unity gain buffer was a very nice sounding one as well,

It also was completely noise-free.

I updated the schematic, it was actually with a gain of around 4, but now it's with a gain of 6 using a 4.7K feedback resistor.

It is in a mesh metal chassis with a toroidal transformer about 4 inches away from the amp module and the power supply between the transformer and amp.

The thing that I'm curious about is that there was absolutely no noise when it was configured as a unity gain buffer and there is quite a bit of a buzz now. Any ideas how that happened?

 

[amb]

There is a difference of almost 16dB between unity gain and gain of 6. Any amount of inherent noise, as well as interference (most likely from the transformer, but you could also have other sources of RFI and EMI) are amplified by the gain of the amp.

4 inches between the transformer and the amp is not a lot of distance especially if the transformer is sizable or if it's an EI-core unit. You can try turning the transformer so that the mounting orientation is changed, to see if it makes any difference.

Also, is your mesh metal case grounded? And how is it grounded?

 

[FallenAngel]

I did a little test today. I took the amp module, rca jacks, pot and headphone jack all out of the case, ran a 3ft power cord to it and had a listen and surprise, surprise, there is exactly the same buzzing noise in the circuit. I then rebuilt the entire thing in a slightly different layout with the same results

Now I'm really frustrated and confused to where this noise could be coming from and why I only hear it when the amp has gain.

 

[amb]

You did not describe the grounding scheme. Bad grounding causes noise. Also, some pictures might help.

 

[MisterX]

PICTURES!

 

[FallenAngel]

Photo of basically how it looks when not in pieces.


PSU is basically 2 TREADS with the a +/- tied together forming a ground.

All grounds go to this point: headphone jack, pot, RCA input and single point on the amp board (there isn't much grounded, just a pair of 1K resistors).

 

[FallenAngel]

Hey guys,

After banding my head against the wall one too many times, I've decided to skip the 2-channel version with the transformer being very close to the amp and just go ahead with the 3-channel version. I'm also thinking of switching the BD139 to MJE243 because the buffer built above was quite a bright one and these should be a little less bright.

This is what the amp schematic looks like:

 

[majkel]

Why don't you build a classic symmetric PSU, instead of messing around with all that ground channel and virtual ground stuff?
What you need is:
- center tapped transformer or with two equal secondary windings (better), 12~17V AC output
- one or two (better, only when you have double secondary winding transformer) Graetz bridges
- some 'lytics
- 7812S and 7912S chips
Sorry for not providing the schematic ATM but I'm in work, away from my EDA software now.