Acoustic 270 - trouble in preamp

[Drummer4gc]

Thanks for your thoughts. I probed the preamp for DC voltages. Many of them are within 5-10% of the spec values, but a few are off by more:

Q102C: 19.3v (should be 16v)
Q105E: 25.5v (should be 16v)
Q106E: 1.6v (should be 25.5v)
Q113C: oscillates from 0-9v (stated as 4v on the schematic, but it's the oscillator so I'm guessing this is fine)
Q115C: 25.9v (should be 30v)

I believe the 5MHz is present on the preamp input as well, and on the power amp signal. I am not sure how to check the power supply using my scope, and I don't want to break it, so some advice would be well appreciated.

I'll take a look back around Q106, as it seems the most off.

[Enzo]

Your scope probe can probe anything. If the preamp runs on say +40v - I forget without opening the drawing - no reason you can't scope the +40 supply just like you would to look for ripple, except this time we are looking to see if the 5MHz is riding on the supply.

I don't care so much of the collector of Q102 is 16 or 19 volts, it is very roughly half the power supply.

But emitter Q106? 1v instead of 25v tells us the transistor is not conducting. What is on the base? Is there good supply on the collector?

[Drummer4gc]

Base is also at 1.6v, with a solid 40v on the collector.

I removed Q106, which gave weird readings on my cheap little component tester - either just a diode (C-B IIRC) or two diodes, but it never registered as a transistor, so it may be shot. Will it be picky about an NPN replacement? Can I chuck a 5088 or 3904 in there and expect decent results?

[Enzo]

Better than no results, eh? What voltage is on the base pad with the transistor removed? The 220k and 470k resistor ought to make about 27v if the supply is 40.

[Drummer4gc]

Yep, base pad is at 26.2 with transistor removed. I replaced it with a 2n3904, and got 26.2 on the base, 25.9 on the emitter. Not sure if this is a legit replacement for the original transistor though.

Lots of extra noise/sound happening now when the volume starts to get turned up. And, with the fuzz switch on, it starts to oscillate when the volume is turned up.

[Enzo]

And quite likely more is wrong with the amp, but at least we now have a reasonable voltage at that stage so it ought to pass signal better.

[Drummer4gc]

There is definitely a larger signal coming out of the preamp, which is great - glad to be on the way!

The loud noise I'm hearing seems to be motorboating, so it sounds like I ought to start checking/replacing electros in the preamp.

The 5MHz noise I'm seeing on the scope seems to be everywhere - preamp input, preamp output, speaker output (no load), preamp power supply, across the main filter cap. I'm wondering if it's oscillation due to replacing the output transistors with modern MJ15003s. Looks like this has come up for other folks trying to repair 270s and 370s, but I haven't seen much of a solution reported online. I still have 6 of the 8 original output transistors in good shape - I'm wondering if I ought to use them as the paralleled outputs and stick 2 MJ15003s in as the drivers to see if that might give a better signal.


EDIT: Tried the original 2n3055Hs as the outputs with the MJ15003s as drivers - still got the 5MHz noise. Then, I tried only the original transistors as drivers and outputs, leaving off the third output on each side - no noise whatsoever. Playing with a guitar sounded pretty clean too.

At this point, I either need to figure out how to make the MJ15003s play nice or source two more original transistors. Any idea if these are legit and appropriate? - https://www.electronicsurplus.com/rc...con-p-n-2n3055

If 2n3055Hs are just not possible to find anymore, are there thoughts on how to make the MJ15003s work? Would love to go this route if it's not too much reengineering, since I've already got them.

Thanks for all the help!

[g1]

Your linked parts seem legit. This guy bought and used them 8 yrs. ago: http://music-electronics-forum.com/s...l=1#post192787

[olddawg]

My 2 cents. Have you completely split the amp between preamp and power amp? If components are bad or drifted who knows what kind of real circuits you have going. I would normally start by creating a functional power amp since that’s where the money is. If you split the amp and have not disconnected the grounding scheme, it should not be that complex. Bipolar output transistors are rated for current, voltage, beta, etc. The junction resistance is always the same. Something close but more robust should work. Are you sure you have you basing correct? You have no idea how many repair are botched by modern subs having a different basing.

[Drummer4gc]

Thanks - I split the power amp and still see the oscillation on the scope, no input no load, so I'm pretty sure it's the new transistors not digging the old design. I'm going to see about grabbing a few old RCA 3055s unless there are ideas about how to tame the oscillation from these faster transistors.

[Drummer4gc]

I replaced the 15003s with 15015s and success! No more oscillation!

I haven't been back to measure voltages yet, but it sounds pretty good, so I think I've got most everything whipped. Thank you all for your help with this - and I hope you can help me with one final question. What's the best way to set the bias for these new transistors?

In doing some research, I've replaced the 680R bias resistor (R419) with a 1k trimmer. I see that lower resistance = higher bias here, but I've seen multiple ways for how to set it correctly, including:

-adjust for ~10mv across each transistor emitter resistor. In this case, I suppose I'd be using the collector resistors R422/R423?
-set the trimmer at 1k, reduce until the idle current of the amp starts to increase (at mains?), then back off slightly
-set the trimmer at 1k, inject a signal and increase bias until crossover distortion appears.

What's the best way to proceed?

Thanks again!

[Drummer4gc]

Bump? Just grabbed a Kill A Watt to measure idle wattage draw. Manual says the amp should draw 50W at idle, but from a cold bias it just starts to rise at around 17W. Not sure where to bias it to accommodate the substitute transistors and would love any thoughts.

[Enzo]

You list three methods above. Be aware that #2 and #3 are the same thing. The point at which crossover notch just disappears is also the point where mains current will start to increase. Those two options will result in a cooler setting than the arbitrary 10mv.


My manual doesn't show the 50w or 10mv spec. Can you post or link your manual?

[Drummer4gc]

Thanks Enzo.

Page 2 of this manual shows "standby consumption" as 50W. https://www.dropbox.com/s/9ujlx06zeq...ice-manual.pdf

The 10mv figure I think I've just seen thrown around in various threads as I've tried to do research on this. I can say, there does appear to be some crossover distortion on the scope as I adjust bias up until ~40-50W, but through reading it sounds like biasing this way is likely to lead to a hotter, less reliable setting.

And of course, one more issue seems to be arising as I'm going through the voltages/trimpot settings with a scope. The amp is clipping asymmetrically. It seems to be a preamp issue - see scope shot in earlier post, and I see it on the monitor out. The balance trimmer on the PA board is set for exactly half the b+ voltage, so I don't think that's contributing. Anywhere in particular to check in the preamp circuit that could be responsible for asymmetrical clipping?

[Enzo]

Oh, I did see that, I didn't interpret that to be a bias adjustment.

So they say idle consumption is 50 watts, and you find that there is crossover distortion until 45-50 watts. SO what is the issue, you find that the well biased amp draws just about what the specs say it does. I don't know what is less reliable about it. The point of biasing the amp is to get rid of crossover distortion.

The balance trimmer on the PA board is set for exactly half the b+ voltage, so I don't think that's contributing.

Read note number 6 in the lower left corner of the schematic. It specifically says adjust R404 for symmetrical clipping.