While I'm certainly not going to argue with any of the points you've made the fact remains that the components are bought, paid for, and installed. No sense doing anything else at this point.

I was aware of that fact that only one of the 68uF caps needed to be changed but I'm a little OCD and can't have things I'll likely never see again mismatched. I doubt such a personal problem is uncommon around here.

I haven't had a chance to power up and set the trimmer pots yet as more important things have come up. I shall update though when I have something to do so with.

Die_Bogener wrote:

There is no need to use 2 of the 68uf caps on the one board. Only 1 of the two caps is responsible for the sound of the woofers. The other one is used for linearisation of the load for the amplifier.

Sorry to disagree.

But this is in my opnion not true. If you look more carefull at the filter, then you will see that the first filtering (2nd order) is done by L1 and C1.

Then the second stage L2 and C3 (combined with R1) is the second 2nd order filter.

In essence 2 2nd order filters combined makes it a 4th order. Only B&O has chosen to choose 2 different cut-off frequencies. And that's done to linearise the frequencies around the 700Hz.

Every filter has it's overshoot around the cut-off frequency. So, by choosing to different cut-off frequencies you can linearise that overshoot.

It has nothing to do with the load for the amplifier.

This kind of filtering does B&O also in it's active filtering with active speakers like the bL6000 and BL8000. Instead of using standard calculated capacators and resistors in these 4th order active filters, B&O uses modified values to achieve the same linear behaviour around the cut-off frequency.

Wow, following this discussion with much interest guys!

I agree that all electrolytic caps on the crossover boards form part of the filtering.

Martin

Well sorry for the lack of updates, not that I think anyone is on the edge of thier chair to hear about whats come of this project... but some things came up in life and I had to box the parts up for a few weeks.

This evening I soldering in the last capacitor I had been waiting on, I referenced my photos of the untouched board and checked it against my work to make sure each capacitor was correct and not reversed. All was well, so in my case anyway the PCB was labeled correctly.

I attached test leads across the emitters of TR27 and TR32.

I plugged the amplifier in and kinda stood back. Nothing blew up, but the power LED was red.

I figured it needed a source to turn on, so I attached an ipod as an input.

I plugged the amplifier back in.

The light turned green for about one half second, followed by smoke, ozone smell, and... flames being shot off R99 and R100. The same resistors that were burnt up last time.

I'm so sad. Does anyone have any advice as to where to look for the problem?

Sorry for saying this, but I wonder why you are surprised, - you haven't repaired anything yet!

Please let me copy my post of aug 22nd:

"Anyway, I think, I would have diagnosed and repaired the bad amplifier module before upgrading and
replacing a lot of components.
... And in case the amplifier module turns out to be a complete write-off, you've lost money and time too."

Søren Mexico posted something to that effect as well.

New capacitors will not fix a blown output stage. - Even shiny ones.
Repairing a DC-coupled complementary power amplifier can at times give even trained tech guys a good run for their money.
In short; Replace the output stage transistors and their emitter resistors, the driver transistors and all other bad components.
The latter being the problem part.
The constant current generator will often have suffered and the idle current trimmer series resistors etc. could also have had it.
When you think, you are done replacing burned components, power the amplifier up using a dual DC-voltage bench powersupply with
current limiting (set to almost nothing initially) and monitor the idle current and the center (output) rail DC-voltage to catch eventual
remaining faults without blowing everything up again.

Martin

I'm not too worried about the small amount of money put into the amplifier components. 90% of the monetary investment was in the crossovers, and ultimately if I can't get the amplifier working I will run the towers bypassing the amplifier.

As for finding the actual problem I fear quickly this will fall outside my scope. My diagnostic equipment is limited to a digital multimeter.

I'm not sure which transistors are output vs. driver.

I'm also not sure how to go about testing any of the transistors to evaluate which ones are or could be bad.

OK so please correct me if I'm wrong but TR25-27 (SC3281) and TR30-32 (SA1302) are the output transistors? Initial investigation looks like both of those transistors are NLA from Toshiba and difficult to source unless you want the cheap Chinese knockoffs. (I don't)

Which are the drivers?

What is the constant current generator?

I have spare emitter resistors to replace R99/R100, everything was ordered in double quantity to give the other amplifier the same upgrades.

Sorry to intrude, but do you have any solid notions in electronics? You seem at a loss here 🙁

No, I'm sure I've said as much earlier in this thread.

I have an interest, learn quickly and have steady/capable hands. I knew I could handle the recapping of the crossovers and that is what I originally set out to do. That part is done and sounds great. I figured I would give repairing the amplifier a shot too, and that didn't go so well as to say I'm right back where I started.

I assume the transistors need to be removed from the PCB to test across the leads. I don't know what to test for, and / or if the test can be done with a DMM or if I need equipment I don't have.