Hello fellow DIYers.
This is going to be a long text but bear with me. 😱
Recently I bought a pair of Beag HEC 45 studio monitors. Really obscure to most people here probably, 3-way active monitors from the late 80s, but I've been looking for a pair for ages and once a pair surfaced, I jumped on them immediately despite them being in poor shape.
The most serious issue is that the amplifier in one of them is dead. What's worse is that someone had a go at repairing it and did a horrible job, basically destroying one of the output cards. A bunch of components broken, missing, replaced with wrong parts, half the tracks on the board gone, etc...
The amplifier's model is Beag EBC 902. It's a 3-way actively XOd amplifier with 4 identical amplifier cards inside. One for the tweeter, one for the mid, two bridged for the woofer. The high and mid cards are fine, the woofer cards are the problem.
The schematics and parts list are attached.
Right now both low modules have heavy distortion. Starting with the first board, I replaced all the capacitors, the TL081CP op-amp because it was missing. Outputs originally were 2N3055 and BDX18, BDX18 is no longer made so I used 2N3055 and MJ2955 paired. All the driver transistors were also replaced, as well as a bunch of resistors and zener diodes. The amplifier was distorting but worked. It almost sounded like a low bias or something, the sound was better at higher levels. I'll bring my O scope home from work so I can measure.
I had a crappy test driver connected to test things. I was hooking the amplifier up to an input and accidentally overloaded one of the inputs. D9, R45 and R47 went up in smoke. I replaced them and also T17 because it's in that path and they blew again so I took out the MJ2955 I had put in and it also measured shorted. I replaced all the components again and put in one of the BDX18s that I still had that was good. Turned it on and it looked like there was a huge amount of positive DC on the output as the speaker was pushed out. Turned it off and took out the BDX18, put in another MJ2955. Turned it on, R47 went up in smoke. Right now when you turn the amp on, it smokes R47. Output transistor survived this time. There is sound but it's still distorted.
I need to bring my scope for further tests but I wanted to put this thread up as this is the most complex repair I have ever done and I will probably need help with this.
Also what is this topology? The front end of it seems pretty complex.
Thank you all for reading 🙂
View attachment 710857
View attachment 710858
View attachment 710859
This is going to be a long text but bear with me. 😱
Recently I bought a pair of Beag HEC 45 studio monitors. Really obscure to most people here probably, 3-way active monitors from the late 80s, but I've been looking for a pair for ages and once a pair surfaced, I jumped on them immediately despite them being in poor shape.
The most serious issue is that the amplifier in one of them is dead. What's worse is that someone had a go at repairing it and did a horrible job, basically destroying one of the output cards. A bunch of components broken, missing, replaced with wrong parts, half the tracks on the board gone, etc...
The amplifier's model is Beag EBC 902. It's a 3-way actively XOd amplifier with 4 identical amplifier cards inside. One for the tweeter, one for the mid, two bridged for the woofer. The high and mid cards are fine, the woofer cards are the problem.
The schematics and parts list are attached.
Right now both low modules have heavy distortion. Starting with the first board, I replaced all the capacitors, the TL081CP op-amp because it was missing. Outputs originally were 2N3055 and BDX18, BDX18 is no longer made so I used 2N3055 and MJ2955 paired. All the driver transistors were also replaced, as well as a bunch of resistors and zener diodes. The amplifier was distorting but worked. It almost sounded like a low bias or something, the sound was better at higher levels. I'll bring my O scope home from work so I can measure.
I had a crappy test driver connected to test things. I was hooking the amplifier up to an input and accidentally overloaded one of the inputs. D9, R45 and R47 went up in smoke. I replaced them and also T17 because it's in that path and they blew again so I took out the MJ2955 I had put in and it also measured shorted. I replaced all the components again and put in one of the BDX18s that I still had that was good. Turned it on and it looked like there was a huge amount of positive DC on the output as the speaker was pushed out. Turned it off and took out the BDX18, put in another MJ2955. Turned it on, R47 went up in smoke. Right now when you turn the amp on, it smokes R47. Output transistor survived this time. There is sound but it's still distorted.
I need to bring my scope for further tests but I wanted to put this thread up as this is the most complex repair I have ever done and I will probably need help with this.
Also what is this topology? The front end of it seems pretty complex.
Thank you all for reading 🙂
View attachment 710857
View attachment 710858
View attachment 710859
Attachments not working.
You can't keep tossing Silicon into a hole. You need a dummy-load, voltmeter, and either a current-limited power supply or two 100r 10W resistors.
You can't keep tossing Silicon into a hole. You need a dummy-load, voltmeter, and either a current-limited power supply or two 100r 10W resistors.
I have a multimeter and a scope. Just ordered some 8 ohm 40w resistors for a dummy load as well as 100ohm 10w resistors. The 100 ohm resistors are for putting in series with the power supply, right? The amp has two dual polarity power supply rails, one for the output section and one for the input section. Should I limit the input section power supply as well?
The attachments do indeed seem to not work. I've re-uploaded them here:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Okay, I've now connected the power supply for the power amplifier section through 100ohm 10w resistors. With a 1 khz input, this is what the output looks like. CH 1 is output, CH 2 is input.
An externally hosted image should be here but it was not working when we last tested it.
Looks like a classic case of crossover distortion (low bias) but that could easily be influenced by having 100 ohm resistors in the supply.
I would turn the bias preset to minimum, then assuming all seems good run the amp on full voltage and re bias correctly. If no figure is given then aim for around 9 milliamps through R51/52 (no signal and no load attached). That's easily checked by measuring the volt drop across the resistor.
I would turn the bias preset to minimum, then assuming all seems good run the amp on full voltage and re bias correctly. If no figure is given then aim for around 9 milliamps through R51/52 (no signal and no load attached). That's easily checked by measuring the volt drop across the resistor.
They come from Hungary in the late 80s 😀
I removed the outputs for now to be safe. With the outputs removed, the graph looks the same. It doesn't look like crossover distortion, the top half of the wave is completely missing 🙁
I measured on both sides of R36 and emitter of T9 and the same graph shows up there as well.
I removed the outputs for now to be safe. With the outputs removed, the graph looks the same. It doesn't look like crossover distortion, the top half of the wave is completely missing 🙁
I measured on both sides of R36 and emitter of T9 and the same graph shows up there as well.
Crossover distortion is the little kink in the blue sinewave around the zero point. I assumed that was the amp output (and wondered what you were showing with the other trace).
If the other is the amp output then you have big problems... but problems that should be easy to fix. Simple DC tests of current and voltage should show where its going wrong. First step would be to check those 0.33 ohm resistors and the output triples (output transistor, driver, pre-driver and related resistors)
If the other is the amp output then you have big problems... but problems that should be easy to fix. Simple DC tests of current and voltage should show where its going wrong. First step would be to check those 0.33 ohm resistors and the output triples (output transistor, driver, pre-driver and related resistors)
Interesting that there's crossover distortion present, it's a 1 khz sine wave and the source is a late 2008 Macbook Pro.
I'll investigate further down the signal path and post what I find.
You would have to look at the signal source as a separate 'issue'. First try it without connecting to the amp and see if the source is pure.
Concentrate on the amp first... all the clues should be there in the voltage measurements.
Concentrate on the amp first... all the clues should be there in the voltage measurements.
If in doubt, I'd put my bets on a shorted T1. Ask the diode test function what it's got to say on the pn junctions of T1, T3, T5 and T7 in particular, and check whether voltages on T1, T3, T24 mirror those on T2, T4, T23.
The complementary input stage is somewhat reminiscent of the bigger Rotels of the time (though its compensation isn't). The 3-stage CFP output is not but rather an upscaled version of what would commonly be found on European amps in the mid-late 1970s. (A CFP output can get closer to the rails than an emitter follower job, and power transistors with high voltage ratings weren't easy to come by in those days.) The transistor types used would also be more typical for that time.
But man, those germanium-era transistor symbols sure make the schematic hard to read. I thought nobody had even used these any more after some time in the 1960s.
The complementary input stage is somewhat reminiscent of the bigger Rotels of the time (though its compensation isn't). The 3-stage CFP output is not but rather an upscaled version of what would commonly be found on European amps in the mid-late 1970s. (A CFP output can get closer to the rails than an emitter follower job, and power transistors with high voltage ratings weren't easy to come by in those days.) The transistor types used would also be more typical for that time.
But man, those germanium-era transistor symbols sure make the schematic hard to read. I thought nobody had even used these any more after some time in the 1960s.
Last edited:
Thanks for the reply! I will check T1, 3 5 and 7 tomorrow.
Here's an interesting thing I found: I tried to see wether the issue is in the front or rear end so I measured where the two stages connect - R36 and R37. The waveforms look identical on both polarities. Why could this be? This is with a 1 khz input signal.
Here's an interesting thing I found: I tried to see wether the issue is in the front or rear end so I measured where the two stages connect - R36 and R37. The waveforms look identical on both polarities. Why could this be? This is with a 1 khz input signal.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
It's the job of the bias circuit to keep a constant voltage between these two points, so no real surprise there. I'd still say your problem is somewhere in the frontend.
Have you gone through all the transistors with the diode test (at power off), just to weed out any obviously shorted ones?
Have you gone through all the transistors with the diode test (at power off), just to weed out any obviously shorted ones?
I pulled out and measured transistors T1, 3, 5, 7, 9 and they all measured good with a multimeter. I only had a multimeter on hand and used diode check as my Ebay transistor tester kicked the bucket and the new one is yet to arrive. So it's not certain that the transistors are 100% good but there were no obvious shorted or open transistors. I even tried a different T1 just in case and it made no difference. As far as voltages on the transistors go, have a look at the waveforms. Output of the amp on the top, measuring point on the bottom. The first pic is measured at emitter of T2. The second pic is measured on collector of T1, so the same point, different polarity. Would this suggest faulty T 5 or T7 maybe? I don't have any BC214 to try and replace them at the moment. I can take more measurements tomorrow 😉
Thanks for the help!
Thanks for the help!
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Does the output waveform look alright with no load attached ?
Is the voltage correct across D3 and D4 (zeners).
Are the base/emitter voltages correct for all the transistors. All should be in the 0.6 to 0.8 volt region apart from T13 and T14 which should be near zero but be careful, one slip of the probes an it all goes bang.
Is the voltage correct across D3 and D4 (zeners).
Are the base/emitter voltages correct for all the transistors. All should be in the 0.6 to 0.8 volt region apart from T13 and T14 which should be near zero but be careful, one slip of the probes an it all goes bang.
Alright, we are in business! Started poking around with the multimeter and measuring voltages and shorts in the input section and nothing seemed wrong. So I started looking at the second half of the amp and noticed a short between the base and collector of T15, measured base of T15, got 30 something volts. Pulled it out and sure enough, dead short between collector and base. Put in a new one and the output is now perfect! There is some crossover distortion still, but we'll have to measure that again once the 100 ohm resistors are out and the outputs are back in.
Can you guys please explain to me how to set the bias?
Can you guys please explain to me how to set the bias?
- Status
- Not open for further replies.