Hk Actor A112 transistors blowing

[mr funky]

Hello all, I hope you dont mind me posting a help question as my first post.

but im at my wits end with this, and its driving me nuts..

 

I have an older actor system which up till now has been faultless.

right at the end of a gig,

a sax player decided to unplug his phantom powered clip on mic

with out muting and there was a loud earth clunk through the pa

I did not think anything of it till I used them again.

and found one of the A112 tops would not come on

powered up fan worked but no o/p,

after checking the f1 4A had blown and T2, mj11015, and T3 mj11016 had blown

 

I replaced F1 T2 T3

and the caps C3 C21 ( bulging abit)

and some of the bc546b's as read in another forum to change them same time as the MJ's

also 4 big caps on the power board ( again bulging a bit)

 

set the Bias as per manual

turned on and it was ok for about 2 mins but when shouted down the mic it blew all 4

transistors on the board.

 

I replaced all 4 MJs again, then put big resistors across the fuse terminals checking for heat

when turning on, ( these then would blow befor the transistors if there was a fault and the heating up would give me a chance to turn off)

all seemd ok so fuses back in set the bias,

I took the tops to a band practice and it was fine, but at the next practice after about 10 mins

F1 T2 and T3 decided to go again, no bang or nothing just off.

 

so now im thinking the bias pot could be faulty but that seems fine I have checked

all transistors

all resistors

all dioeds

all caps

checked the tracks

checked for shorts

checked wiring

even re soldered the MJ tran screw bases for good earth contact.

thats where im up to now.

before I turn it on again and blow more transistors

I thought I would post here and ask if anyone has experienced and found any intermittent probs

with the A112 or any ideas ??

Any thoughts greatly welcomed and very much appreciated.

 

if this is in the wrong section just shout and I will move it

Thanks

[laolu]

Assuming the drivers (esp. woofer) weren't damaged, I may proceed like this:

 

I would first disconnect the power amp from the pre-amp (if possible), then: can you check if the voltages are all correct? Do you have a scope? then you could check if the amp is maybe oscillating? If nothing gets hot and the voltages look fine, let it on for a while. Then - if you got one - add a function generator to the input of the power amp and slowly turn up volume, checking with a scope what the output does on a 4Ohm resistor. Especially also check for DC on the output. If you got DC or clipping before you get a "decent" output signal, the fault is still in the power section of the amp. Do you have a second (good) one to compare it with? Good luck!

 

- end of wisdom -

 

Norbert

[mr funky]

correction for the first post the speaker is a AT-112A

Thanks very much laolu,

yes I have the other side I have been doing comparison checks against, again all seems fine.

checked the 12" speaker the horn and cross over all is good.

tho the black wire going to the BMS horn

was very loose was off when I took it out ( tho horn was working so must have been connected)

 

I can lend a friends scope and sig gen, that seems to be the next stage,

I just wondered if anyone had any experiences the same as mine and could give me some ideas.

cheers again

[mr funky]

for future ref, incase anyone else gets this problem,

the fault was a dodgy BD679,

this is the one under the heatsink, there is 1 on each side

( my faulty one was number T9 on the schematic)

this is part of the drive for adjusting the bias

so when setting quiscent current adjust to 6.6mv to 8.8mv

with trim pot TR1, it would just be pointless as T9 altho testing ok

when it acts up would throw TR1 out,

So replace the Bd679 when changing the pair of O/P trannys mj11015 and MJ11016.

 

So if trannys on one side blow I think as a good overhall for that side

you would need..

1X Mj11016

1X Mj11015

1X Bd679

2X BC546B T7, T8 ( hk recomend replace these as well)

2X caps C2 and C3 if bulging on top abit.

4X the big power caps on the power board again if bulging on top abit.

 

also check the Zero ohm resistors over the O/P trannys you will have to take one side out

and the diods BY500.

 

schematic here....

http://diagramas.diagramasde.com/audio/Actor_service_instruction.pdf

[mr funky]

AAAAARRRRR ....F^&k

 

they have gone again !!

 

any ideas any one ?

[Jivemaster]

Have you checked all the insulating materials? New mica/silicone insulating washers between hot things and their heatsink and between heatsink(s) and chassis if needed. Does the design need heat sink compound? Have you used the right stuff? -it takes so little swarf to puncture the insulation from device to heatsink.

[mr funky]

Have you checked all the insulating materials? New mica/silicone insulating washers between hot things and their heatsink and between heatsink(s) and chassis if needed. Does the design need heat sink compound? Have you used the right stuff? -it takes so little swarf to puncture the insulation from device to heatsink.

 

Cheers for the fast reply.

yep done all that

also the heat sink is not grounded at all but suspended, as the transistor collector ( case) is directly connected to it with no mica insulator, nothing touches other than the case bolts, I even insulated the base and emitter legs to make sure,

cheers

[peddrums]

This sounds all too familiar; I used to have an Elias PX system which would fairly regularly blow its output transistors but fixing the poorly amp was never a problem...

 

Then one day one of sub's amps blew so I fixed it and it blew before I even started biasing it, so I fixed it again... Blew within minutes. I probably had 5 or 6 good goes at it and gave up, never did find the problem.

[DrV]

Have you checked the bias? The service manual says set it to give 6.6-8.8mV across the emitter resistors. This is twice what is recommended for the Elias which is essentially the same amp (but the speaker may be a different impedance - can't check that right now). Even though the SM for the Elias says 3.3mV I always set it a little lower (about 2.5-2,7mV) This is plenty to prevent crossover distortion which is what it's there for. In your case I think I would just wind it up enough to remove the crossover distortion whilst also checking it isn't higher than 6.6mV.

 

Also it seems obvious but is the fan running and the correct way round? Airflow should be in through the fan and out over the heatsinks and then exhaust through the grilles at either end. And if you're testing it on the bench make sure that it is resting on the PCBs so as to form the tunnel between them that the airflow depends on.

 

Then there's always the possibility that your speaker has a shorted turn, but you can't really test for that other than with a proper impedance tester like a WT3

 

I've fixed loads of these and the only time I had one back it was because the other pair of transistors had blown, not the ones I had replaced.

 

HTH

Dave

[mr funky]

Hi dave and ped,

Yes I always check the Bias, only this time

I put two meters on both sets of bias resistors ( each side of the board)

tests with the amp on bench no load, vol at 0, turned on.

the left hand pair ( side that did not blow) was a steady 7.5mv

but the side that had blown once set to 7.5 mv after about 20 30 seconds the mv would start to climb up slowly

( if I set it to 6.9mv it would start to go down !.)

so I kept tweeking it back 7.5 after about 20 mins of tweeking it would become stable.but then start to climb a bit but more slowly

so..

But using the 10k pot , to get it to 7.5mv the pot was hard over and if you even moved it a gnats pube it would leap by 4 or 5 mv, so it is a pain to set right

I think the pots are far to high to accurately set the bias, so I tried a 4.4k pot I had, and got alot more scope on it and was able to set the bias so much better, ( so luckly hard over must have been at the begining of the pot the pot was still reading 10 k ok)

and now its holding the 7.5mv

connected the speaker, drove it for a few mins while watching both the meters and when the volume was backed off to 0 the both settled to 7.5 again.

 

so could have been a dodgy bias pot after all that as 10 years of toll would have had an effect no doubt.

and they are prob not the best ones to use anyway.

so I am going to swap them for 10 k 25 turn pots to get a much more accurate steady state.

like these.

http://www.ebay.co.u...#ht_1851wt_1093

 

 

Al keep yall posted

 

( word of warning if you are taking out the amp a few times, try not to bend the ribbons too much ! as the wires joining the board will start to snap on the sholder

with all the movement.)

[DrV]

You may already know this but if the bias is unstable over the long period you should check the transistor which is trapped under the heatsink, and it's thermal contact with the heatsink. As the output transistor warms up its purpose is to detect this and reduce the bias accordingly. Haven't got the schematic to hand but I'm sure you know the one I mean.

 

If that transistor has a dodgy joint or is itself intermittent then that will quickly allow a large shoot-through current which will destroy the output.

 

And you're right about those nasty interconnects!

[mr funky]

You may already know this but if the bias is unstable over the long period you should check the transistor which is trapped under the heatsink, and it's thermal contact with the heatsink. As the output transistor warms up its purpose is to detect this and reduce the bias accordingly. Haven't got the schematic to hand but I'm sure you know the one I mean.

 

If that transistor has a dodgy joint or is itself intermittent then that will quickly allow a large shoot-through current which will destroy the output.

 

And you're right about those nasty interconnects!

 

 

Thanks v, yep been there also and replaced that.

its doing its job as there is only one for both sets of transistors and the other side is stable

[KevinE]

check and double check the grey ribbon cables to the pcbs, they can fracture after a repair leading to all sorts of intermittent mischief

[richardash1981]

You may already know this but if the bias is unstable over the long period you should check the transistor which is trapped under the heatsink, and it's thermal contact with the heatsink. As the output transistor warms up its purpose is to detect this and reduce the bias accordingly. Haven't got the schematic to hand but I'm sure you know the one I mean.

 

If that transistor has a dodgy joint or is itself intermittent then that will quickly allow a large shoot-through current which will destroy the output.

Thanks v, yep been there also and replaced that.

its doing its job as there is only one for both sets of transistors and the other side is stable

I have grave doubts as to there being a single bias compensation transistor for both channels of a two channel amplifier, and the schematic shows T9 and T13 in the bass section (for the two halves of the bridged amplifier) and the same on the top section. So the top section you are working on has two bias generator transistors. Unfortunately, if one fails, then because the design is a DC coupled bridge, it's quite likely that all four output devices will fail in quick succession - giving the voice coil a major hammering in the process.

 

Certainly I'd change the trimmers, and check the associated fixed resistors are present and correct.

[DrV]

I think that's just semantics (or is it grammar) - there is, of course, one bias transistor for EACH set of output transistors, not BOTH sets