barbetta 3105 schematic

[Jazz P Bass]

I just reread the posts & something about post#5 struck me as odd.
Quote: "Also the MJ15003 and MJ15004 output devices both have -5vdc on their emitters and their bases both sit at -4.65vdc"
Does this mean the drivers are blown?
Both bases should not be at a negative voltage.

[lowell]

I agree that's why I posted it. One base should be positive. I still need to investigate the LM391's.

[Jazz P Bass]

Well what are the bases of the drivers at?
You very well may have to yank the LM to prove it out.
(I did find a supplier (Little Diodes, in England) that has stock of the LM391N-100's. 2 week delivery.
I would like to see a trace of the board.
There are two diffent kinds of transistors & an N & a P mosfet.(what are they doing?)

[rf7]

Even without schematic you already have some important measurements do do.

Such as rectifier diodes, filter caps, power transistors, unloaded transformer voltages, case to heatsink insulation, etc.

I got it. I found 2 driver transistors (TIP107 and a TIP102) that measured low impedance between the C-E junctions, in and out of circuit. I had already checked all the other stuff that you mentioned before I posted.

By the way it was a Sona-Elan SE-41C amp, if anyone cares.

Thanks!

[Jazz P Bass]

Do you recall what base voltage you had on the TIP drivers? (for the OP)

[lowell]

Ok

TIP102
b: -3vdc

TIP29
b: -5vdc

So it seems the issue is definitely before the drivers then? The LM391N?

I am going to post the voltages for this IC.

[lowell]

So I don't have a good setup here at my new space. The best I can do is put this in writing for now. Pins 5/6 are soldered together on one LM391 IC. They go to the trim pot as well as to a 2k resistor, 270 resistor, a 1n47 anode. The 2k/1n47 are in parellel and the other side of them connects to a transistor. Can't find any info on this transistor. It says P8P20 and M916 on it. The M looks like a Motorola logo. Pin 8 of this LM391 has the same value components on it as well and they go to a transistor-looking device with P8N20 R912 on it. Pin 2 of the P8P20 goes to the HighV negative rail. P8N20 Pin 2 goes to HighV postive rail.

I hope this makes sense and that someone can tell me what these devices are and what their purpose is.

[Jazz P Bass]

Have the TIP drivers been tested?
If they are good that leaves a bad bias transistor inside the LM391N.
At that point I would pull the IC & see what is up with the voltages.
As to the parts layout:
The P8P20 is a P channel mosfet.
The P8N20 is an N channel mosfet.
So you have an LM 391N IC.
Two mosfets.
Two TIP drivers.
That leaves the two 2N6474 & 76 transistors.
What are they hooked up to?
And what does the 'other' trim pot do?
A down & dirty scanned drawing would really help.

[lowell]

The TIP junction tests do look good.

The 2n6476 emitter goes to the MJ15004 as well as a 1/2w 470ohm resistor. That 470ohm is connected to another 470ohm which is connected to the 2n6474 emitter as well as the MJ15003.

If I had to guess, the 2n6476 and 2n6474 are limiters. The TIPs are the drivers. So that seems like all the transistors are accounted for. Make sense?

Could it be that the two P8 mosfets drive the tweeter? Not sure how the tweeter is driven from this. However these mosfets DO connect to two 22ohm 3watt resistors which then connect to the tweeter. Which to me looks like those 22ohms are the ballast resistors for the tweeter PA.

The other trim pot is for the OTHER LM391. Just like the first one, this pot is in series with 2 1/2w resistors and between pins 5/7 of LM391. The two 1/2w resistors on here however are a 1.5k and an 8.2k. The junction of these 2 resistors goes to the BASE of TIP29.

Apologize for doing things this way, but I'm pressed for time tonight and drawing it out will take too long. I will attempt a drawing as soon as I have more time. (just became a Dad 3 weeks ago! )

[J M Fahey]

(just became a Dad 3 weeks ago! )

Hey!!! That's GREAT !!!!

Congratulations !!!!!

[lowell]

Ok so I was kinda off a bit. The mosfets are driving the tweeter and the TIPs are related to the tweeter PA as well. I THINK they are acting as limiters? Either that or related to bias in some way? Maybe tells the LM391 to keep those mosfet gates a certain voltage apart? The 2n transistors are probably the drivers for the woofer PA. Here's what I drew out for the tweeter PA thus far. BTW the tweeter doesn't work either. There IS teeny tiny bit of sound coming out of it, but really it ain't workin' right.

[Jazz P Bass]

What is Pin #6 hooked up to?
That is the IC internal bias base pin.

[lowell]

It's tied directly to pin5

[Jazz P Bass]

That's weird.
The datasheet implies that a 39K resistor there (between pin 5 & 6) is a good place to monitor while setting the bias to 25ma.
Oh well.
If you remove the ic there are certain 'diode check' tests that you can do to see if the ic is shorted.

[lowell]

I should point out that pins 9-13 are soldered together on both LM391. That seems odd... like Barbetta is eliminating the limiters. That seem right? Finally I am getting ~.4-.7v diode checks between pins. Do I need to be more specific? Nothing is a dead short.