ARCHIVED FORUM -- March 2012 to February 2022READ ONLY FORUM
This is the second Archived Forum which was active between 1st March 2012 and 23rd February 2022
Hi all
I have bought 2 defective beolab penta 1 amplifiers. Needed some of the trim for spares for my working set.
But then I figured I would try to fix them and maybe use them for something. Both powered on, but 1 with orange light right after. The other would go into orange after 30 seconds of green light.
The one with green light I figured I would look at first, and thought it might be a bad solder at first, and I also found what looked like bad solders here and there. I re-soldered the lot, and was quite careful since the traces are sensitive.
Plugged it back in, but now no red light, and after some time, the 40mA fuse blows. I managed to get some quick measuring of voltages done before the fuse blew, and it seemed as the +15V was missing from IC5, -15V was fine. Then the fuse blew.
i can see no shorts around IC5, but something is shorted for sure. The control print is as is, I didn't re-solder that, but I cannot find any shorts with my resolder. Are the 78xx/79xx IC's very sensitive to heat or anything ? Or is there another common failure for the 4mA fuse to blow ?
Is it ok to take the print out of the cooling case while troubleshooting, or will that damage the power transistors if and when it at some point powers on again ?
/Jacob
As you measured that you're missing the +15V, change the 7815.I've had to replace them several times.
Replace also the 33uF electrolytics. That can fail/shortnend
Beobuddy: As you measured that you're missing the +15V, change the 7815.I've had to replace them several times. Replace also the 33uF electrolytics. That can fail/shortnend
But would that cause the fuse to blow ?
In some cases, yes. But the small rectifier next to it could also be the culprit.
Beobuddy: In some cases, yes. But the small rectifier next to it could also be the culprit.
It is just funny that it happened after the resolder, are these components that sensitive to heat, or could they have been bad and the heat just pushed them over the edge ?
anyway, I will try to swap the lm's / rectifier from the other amplifier, and see if that solves it.
My plan was to completely resolder the pcb1 from the Penta 1 amp and use on my Penta 3 amp. The 3 has many cracked traces and solder points, and the 1 looks like new. Also I would swap the transformer over since that is also very rusty on the 3, and I suspect it to cause humming in the circuit.
I have looked at the diagram for 1 and 3 PCB 1, and they look very much the same. Do you know if there are any differences between 1 and 3 regarding trafo and other stuff I have missed ?
Quick update..
The PCB 1 are not identical on the Penta 1 and 3, 99% though, so maybe I will give it a go ahead with the exchange with small modifications, maybe not :)
Rectifier, IC5/6 and all caps around IC5/6 are fine, but it still does not work. When I power it on WITH LM7815, there are 4v at the input pin, there should be 22, and there are 3v at the output pin. LM7915 has -23 on input, and -15 on output, that is fine. If I power on without LM7915, the input has the correct 22v, so something is shorted somewhere, but not the 7915, that works fine in the other amp.
Did some testing, and at first Diode D9 seemed to be dead, and was open both ways, but after de-soldering, it was also fine, but when R12 is connected, the diode tests fails, with R12 liftet, it passes, so have to test further on down the path. The working amp is a great reference, and it passes the diode test with all components connected, so something has been shorted or died during my re-solder, although I have been through all the solder points many times to see if I have created a short, I cannot seem to find one though..
So, replace the LM7815.
PCB mk1 indeed differs from the mk3 version.
The mk1 misses the vertical standing pcb next to ic1 and ic2.
If you disconnect the output pin from the LM7815 and you still have no 22v on the input pin, then the LM7815 is defective.
If he isn't, you will have +15V on the output pin itself.
Beobuddy: So, replace the LM7815. PCB mk1 indeed differs from the mk3 version. The mk1 misses the vertical standing pcb next to ic1 and ic2.
I have switched the LM7815's between the working and non-working amp, and both LM's works fine in the working amp, and both does not work in the defective amp, so it must be something else than the LM that is dead ?
Also, my board has the vertical print next to IC1/2... Could it be a MK2 board ? It has line-in, 1 speaker link input, and red/black terminal input.
Beobuddy: If you disconnect the output pin from the LM7815 and you still have no 22v on the input pin, then the LM7815 is defective. If he isn't, you will have +15V on the output pin itself.
I will try this later, I have tested the LM7815 out of circuit with 5 volts, and it seems to put out 5v fine on the output pin. I know this is not the best test, but at least it is not shorted or totally dead ?
LM7815 is fine, if I lift the output pin, I get 15v fine. When it is connected, the input drops to 5v and output to 4v, and then after some time 40mA blows(have found out that I can measure for approx 10 seconds before blowing the fuse)..
As I wrote, when measuring over D9, there is connection both ways with the diode tester, if I lift R12(connected directly to ground) then D9 is acting normal. I have been measuring and following tracks like a maniac, and I really cannot find any shorts. I have removed IC1/2/3 just to test if one of those had a short to ground, D9 is still open both ways, the only way I can get D9 to act like the working amp, is by lifting R12(for now, still have some other components to lift).
What I am doing, is just lifting components that has connection to +15V and ground, but since I have a limited time to measure with power on, I use the diode tester over D9 constantly after each lifted component.
I am unsure whether I need to find a short to ground, or to the -15V. Are there a better way of finding a short than my method ?
Any bright ideas :) ?
Thanx
Update, I found the issue...
Damn, this was not easy, but I started from the LM IC's and compared all traces against a known amp, working through the board with the DMM. I thought I had done this many times, but today was my lucky day.
Found that solder from one leg of C13 was actually touching the trace just to the right of it, that caused the short !!
So no dead components or IC's just me that was unlucky with the de-solder tool I guess, it must have hit the leg somehow, and when I then did the resolder, it touched the trace just below.
Sorry for all the trouble ;)
Faults like this with no "natural cause" are impossible to diagnose from a distance.Good job !
Martin
Dillen: Faults like this with no "natural cause" are impossible to diagnose from a distance.Good job ! Martin
Yes, and I probably never would have found it if I didn't had the know working amp to compare with :) I had to lift a lot of jumpers and resistors before I came to C10 and C13 and found out that C10 gnd had a short to C10 +15V.
Now I understand why these faults can be so hard to find, I actually was thinking at one point, to just remove all components and clean the board from all solder, and then put in 1 component at a time, making sure that there only was connection where they should be :) Thank god I didn't had to go there, although my plan actually was to use the PCB1 from the Penta 1 amp, for my Penta 3 that has a lot of bad connections on P1, but they are not 100% the same, so I do not know if I could get it working, For now the Penta 3 is working with jumper wires.