Findip MA110

[gurkan75]

I have spent some time troubleshooting a couple of faulty Findip Ma110 microphone preamplifiers. They all seem to measure the same incorrect dc values compared to a working example. I think I have figured the first one out, a faulty transistor 2n2905 (marked red) in the output stage that shorts to 0v loading down the 30v rail through a conducting complimentary transistor 2n1711.


I didn´t have any 2n2905 on hand so I tried a 2n4033 which seemed close enough in the data sheet. The voltages actually line up very well now, but I am wondering if it is a coincidence? The amplifier looks like a push-pull design and my question is, should I also replace the "other" (there is a third) 2n2905 with another 2n4033, matched for Hfe? (Or even match to the 2n1711?)

Any comments appreciated,

Johan

 

[abbey road d enfer]

I have spent some time troubleshooting a couple of faulty Findip Ma110 microphone preamplifiers. They all seem to measure the same incorrect dc values compared to a working example. I think I have figured the first one out, a faulty transistor 2n2905 (marked red) in the output stage that shorts to 0v loading down the 30v rail through a conducting complimentary transistor 2n1711.


I didn´t have any 2n2905 on hand so I tried a 2n4033 which seemed close enough in the data sheet. The voltages actually line up very well now, but I am wondering if it is a coincidence? The amplifier looks like a push-pull design and my question is, should I also replace the "other" (there is a third) 2n2905 with another 2n4033, matched for Hfe? (Or even match to the 2n1711?)

Any comments appreciated,

Johan

I don't see any significant difference between the 1033 and the 2905 that would make me reject the 4033. You may feel more comfortable replacing both, but I seriously doubt you could measure a difference, even less hear it.
This circuit topology makes it very tolerant to semi-conductors variations.

 

[gurkan75]

This circuit topology makes it very tolerant to semi-conductors variations.

This is good news, thank you for your insight.

Johan

 

[ruffrecords]

If my calculations are correct, there is over 40mA flowing in each output stage so each of the four transistors is dissipating over 600mW which is too close to my mind to the 800mW @25C dissipation spec of the 2N44033.

Cheers

Ian

 

[abbey road d enfer]

If my calculations are correct, there is over 40mA flowing in each output stage so each of the four transistors is dissipating over 600mW which is too close to my mind to the 800mW @25C dissipation spec of the 2N44033.

Cheers

Ian

That's correct, I noticed that, but the original 2N2905 has the same rating. I believe the designer took the risk to operate without safety margin. Did he conduct tests to prove experimentally that he was right, I don't know.
Does anyone have reliability reports on these?

 

[gurkan75]

If my calculations are correct, there is over 40mA flowing in each output stage so each of the four transistors is dissipating over 600mW which is too close to my mind to the 800mW @25C dissipation spec of the 2N44033.

Cheers

Ian

Thanks for chiming in. Current through 22R times voltage drop across collector-emitter, is that correct? If so, I am measuring a bit more than 600mW.

The data sheet I have on 2n2905 for instance says 600mW power dissipation but maybe that number is given  without heatsink? There are little individual heatsinks on all transistors in the output stage.

Either way I am finding enough dead transistors to want to change them all in the output section. I know I have some bd135/136 which are rated at 8w (<70C) 45V. The different pinout will put them in positions that makes it hard to mount a heatsink though. Do  you think they'd survive without?

EDIT: I believe the power dissipation of bd135/136 without heatsink is much lower (a tenth of?) than 8w meaning I probably need one.

The 8W is because the transistors I have are Philips, which were rated lower than current production 12W in most data sheets.

Johan

 

[gurkan75]

That's correct, I noticed that, but the original 2N2905 has the same rating. I believe the designer took the risk to operate without safety margin. Did he conduct tests to prove experimentally that he was right, I don't know.
Does anyone have reliability reports on these?

Thanks, I spent yesterday trying to repair these preamps and 4/5 have 1-3 dead output transistors.  It seems unlikely these transistors were the best design choice in the long run. It probably proved itself adequate at least until the DIN-cassette system (v72 etc) with self contained 220v power supplies went obsolete only a few years later? Forgot to mention they are v72 type cassettes.

Also, higher mains voltage today  230+ vs 220 isn’t helping.

Johan

 

[abbey road d enfer]

Actually I think there is no good reason to operate the transistors with such a high current.
This operating point makes it capable of driving a 600 ohm load at +20dBu! Unless you want to record rocket launches, I don't hink it's necessary. I also doubt you will load it with 600 ohms.
What I would do is reduce the idle current, by replacing the 3.3V zener diodes with 3x 1N4148 in series, that would reduce the current to about 16mA.
Beware that the diodes should be oriented in the opposite direction of the zener.

 

[ruffrecords]

Or perhaps change the 4 x 22R to 4 x 43R.

Cheers

Ian

 

[abbey road d enfer]

Or perhaps change the 4 x 22R to 4 x 43R.

Cheers

Ian

Yes, as is often the case, many ways to skin the proverbial cat.