Home Theater Forum and Systems banner

1 - 8 of 8 Posts

·
Registered
Joined
·
13 Posts
Discussion Starter #1
Hi There keen Shacktser Techy's..

I have a problem with my DIY 2ch Cahnnel Amp.
It's the complete kit from Silicon Chip, the Ultra-LD Stereo Amp (Altronics kit # 5155).

I've had it for a while and it's been working beautifully up until a week ago..
It's used to power HF drivers in a bi-amp active crossover setup, so it's never really working anywhere near "hot"..

Symptom: The right channel works perfectly as always. The left channel has no output. Well, that's not completely true. Occasionally, the left channel 'fades' in, and output is at the volume sounding crisp and clear. It might might perform normally for 10 minutes, or sometimes not even 1 minute, and then the output quickly 'fades' off to no output at all..

This cycle might happen again 5 or 10 minutes later, or sometimes it stays no output until I switch it back on the next day :doh:

The DC supply rails from the Power Supply are paralell-ed to each channel, so it cant be power supply.
The Pre- amp is on a seperate board, and I have swapped the inputs to the Power modules and the problem remains on the left channel, so I'm fairly certain it's the Left power amp module..

Any feedback would be appreciated.. I have a signal generator and multi-meters, but no cro. And obviously having built this myself I have shcematics.

Cheers..
Murphy
 

·
Premium Member
Joined
·
129 Posts
I just attempted to find out something about this amplifier but spectacularly failed :rolleyesno:

I have heard of similar problems in different equipment & it often boils down to a faulty capacitor with these symptoms. I'd suggest you first of all try bypassing the input capacitor on the power amp with the faulty channel. This is most likely going to be a standard electrolytic. To test it use another capacitor of approximately the same value & connect through the new cap to the amp side of the input capacitor & with the function generator on very low see if you get a consistent output (+ input through the capacitor & negative to ground).

There aren't likely to be any other electrolytics in the signal path, though i could be wrong on this score. If you have no joy having done this see if you can find a circuit diagram so i can take a look & possibly advise on what things to check with a DMM. Obviously an oscilloscope would have been better but we'll do what we can with what we have got :)

Good luck :T
 

·
Registered
Joined
·
13 Posts
Discussion Starter #3
Thanks for that Event Horizon..

I didn't any other feedback :-( It seems the days of fault finding and component level repairs are over by the majority..

I will try your suggestion,, and I should be able to get you a circuit diagram quite easily.. For copyright purposes I think i might have to send it to you in a PM.

Hopefully it's just capacitor and not anything mounted to the heatsink :-0

Cheers..
Murphy
 

·
Premium Member
Joined
·
129 Posts
I have the circuit diagram :) Not a bad implementation, though i'd have done a few things a little differently. I'll ask & see if it's ok to post it as frankly it'll be a bit of a pain going through something that no-one can see :rofl:

As i suspected the input capacitor is an electrolytic, i conveniently forgot about the capacitor in the feedback loop - far less likely to be a problem but still a possibility.

For some unknown reason the output transistors have no emitter resistors so current sharing will more than likely be problematic (though this isn't the problem you are experiencing) :rolleyesno: As i say, there are a few things i'd change & the above would be first on the list.

Lets see if we can get this circuit diagram up here for starters :)

There is no reason as far as i know why we can't at least link to the circuit diagram etc :D

By the way, having looked at the build of this amp i wouldn't have built it this way either :scratch: The output stage is known as a CFP or Complementary Feedback Pair (also known as a Sziklai output stage), as i say there are no emitter resistors on the output transistors, this would be perfectly fine if there were only one! But with two per rail (not including the drivers - MJEs') they won't share current properly. Also he shows the transistors & drivers & bias transistor bolted to one heatsink - wrong :rolleyesno: With a CFP the bias transistor (in this case an MJE340) needs to be in thermal contact with the drivers only! Any other way will cause standing (quiescent) current instablities over temperature :whistling:

As i say, i'd do things differently :D
 

·
Registered
Joined
·
13 Posts
Discussion Starter #5
Hi Horizon,
I'm glad the link worked :)
Don'y take the photos of the actual cicuit boards too literally, they are protype photos, the final layout is much nicer.. But the circuit diagram is accurate with the final "release" of the kit.. Although not perfect my any means, this amp is (was) a great performer, with a very good noise floor and dynamic range, and I'm keen to get it back in my rack..

Tonight I will test the input capacitor. Which is the capacitor in the feed back loop?? is it the 1000pF that links the base of Q7 with emitter of Q10 ?? forgive my ingnorance!!


Cheers..
Murphy
 

·
Premium Member
Joined
·
129 Posts
The cap in the feedback loop is the 100uf 16Volt electrolytic that is just below the long tail pair, IE Q1 & Q2. If this was messing about the amplifier would go from it's full closed loop gain to unity gain. In other words the volume would drop quite significantly but there would still be some output rather than none. depending on what level you were listening at you might actually think that there was no output from that channel, definately if the working channel was still playing music.

I suggest you check the 10Ω resistor from the -Ve input to ground & also the 1KΩ resistor on the +Ve input to the Q1 base. Other than that have a good look at all soldered joints as it could be something as simple as a dry joint :T

If no joy then we'll have to have a look about with the DMM while it's working :doh:

The 100pf capacitor you were refering to is what's known as the "miller" capacitor or dominant pole. This cap controls the frequency where the open loop gain of the amp rolls off. The amp would most certainly go into severe high frequency oscillation until the output transistors expired if this was faulty or absent - maybe a few seconds :D
 

·
Registered
Joined
·
13 Posts
Discussion Starter #7
Thanks Even Horizon..

That's some good starting point for me.. I will keep you "posted".. I have already inspected for dry joints and breaks in the tracks etc.. So I'm failry certain that is a component issue (Hopefully nothing mounted to the heatsink!!) so I will start a series of tests looking at the areas you mentioned..

Cheers!!
Murph
 

·
Registered
Joined
·
13 Posts
Discussion Starter #8
Event Horizon,

Thanks for your help!! Was indeed the input capacitor... I recall now that I accidentally switched to a live input with the volume turn up over half way and it almost blew me off the chair with fright!! I wonder if that's what caused input capacitor to become faulty.. I might change the other one too while i've got the case opned.. Anyway it's good to have it fixed :jiggy:

Well done..

Murphy
 
1 - 8 of 8 Posts
Top