Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 3, 2016 20:16:31 GMT
After blowing my amp up by connecting the speaker leads totally the wrong way around when I was tired I have decided to repair it [again] with the kind help of Frans our resident genious To save time I took the amp to bits this afternoon and took a picture of the board - no blown fuses and nothing obviously wrong like a burnt out resistor or something? not even a smell of anything that may have blown. I measured the bridge rectifier as per last instructions when I blew it up before and they were pretty much the same. I remeasured all the MAG NPN 2SC5200 & MAG PNP 2SA1943 output devices as per last instructions - results bleow. These also seem to be OK? So I was wandering if the other transistors at the front of the amp may be the problem? They are MPS A92 & MPS A42 high voltage transistors and there are 12 of them. I hope this will save us some time Frans?
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 4, 2016 6:03:03 GMT
Seems to be O.K. Are you sure the fuses measure O.K. ?
The smaller transistors are unlikely to blow when the amp is shorted. In most cases the power transistors, fuses and most likely the BR will give up.
Next step is to measure voltages when powered up. One 'slip' with the probe could cause lots of damage so is a tricky business.
When powered up the following measurements are of interest (all DC).
back lead on ground: red lead on: C of 2SC5200 (need to measure 1 only) + DC power rail voltage somewhere in the region between +60 and +80V ? C of 2SA1943 (need to measure 1 only) 1 DC power rail voltage somewhere in the region between -60 and -80V ? output of the amp (should read 0V +/- a few mV DC) B of 2SC5200 (need to measure 1 only) expect around +0.8V E of 2SC5200 (need to measure 1 only) expect around +0.1V B of 2SA1943 (need to measure 1 only) expect around -0.8V E of 2SA1943 (need to measure 1 only) expect around -0.1V
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 4, 2016 20:25:23 GMT
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 4, 2016 21:31:00 GMT
It looks like the rectifier is still O.K. Just the transistors and fuses may be blown. Maybe even just one NPN and PNP transistor ? With the alu bars the transistors may have held 10 mins or so. Enough to take some measurements.
You have to unsolder each and measure it individually.
The suggested parts seem to be viable options. 4MHz is fast enough... they are used as 'emitter followers' so will still be a lot faster than the amplifier itself. The other specs (gain, voltages, currents etc.) are all equal. I don't think the amp nor speakers nor ears will reach 1MHz anyway. Would recommend to replace all 6 of them at the same time.
No need to worry about the other amp not having the same output devices though. The output transistors are not the bandwidth limiting parts in this design.
Sometimes the driver transistors also could be blown. In case you want to be sure just follow the traces from the B of the power transistors to any of the small transistors and perhaps replace these small ones as well. Could also measure these after the power transistors have been removed.
Skype may come in handy by that time.
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 5, 2016 21:13:56 GMT
It looks like the rectifier is still O.K. WHAT With them measurements and my not so good understanding of rectifiers I would have said it was a Gonna. They are made up of 4 diodes yes? so surely they should not have a reading in 4 directions? For £3 I will probably put a new one in unless you can convince me it is OK Just the transistors and fuses may be blown. Maybe even just one NPN and PNP transistor ? Your Psychic Frans One PNP & one NPN was totally trashed; The others read: NPN Red on C Black on E 1 Black on C Red on E 1 Black on C Red on B 541 & the other 542 PNP Red on C Black on E 1 Black on C Red on E 1 Red on C Black on B 565 & the other 569 Again with these costing £3.50 each it's a no brainer to replace ALL 6. Sometimes the driver transistors also could be blown. In case you want to be sure just follow the traces from the B of the power transistors to any of the small transistors and perhaps replace these small ones as well. Could also measure these after the power transistors have been removed. Skype may come in handy by that time. Do you mean then MPS A92 & A42 transistors? The traces are really difficult to follow, so again for what they cost @ approx £1 for a 5 pack I think the safest option is to replace them all and that eliminates them also. I will unsolder them all first and take readings but I think replacement will be better? Yeh Skype may be handy soon. always better to have two brains on the job - sorry I mean one and a bit brain Thanks for your patience Frans
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 5, 2016 23:12:06 GMT
When you measure the bridge rectifier again with the power transistors out of the way you will find that the BR probably measures O.K. Do that first.
The A42 and A92 transistors that may possible be damaged are TR6 and TR8 but I suspect they have been protected by FET1 and FET2 so see no need yet to replace them all.
What I would do if it were my amp I would put back the 4 remaining good transistors in their positions, thus NPN and PNP in the proper positions (they are in parallel so the exact position left open doesn't mind) Make sure they are mounted on the heatsink, replace the fuses and fire it up shortly see if something bad happens. If not (it just gets warm after a short while) do the earlier suggested measurements. keep the temperature of the transistors low. Switch it off when they become unreasonably warm/hot.
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 6, 2016 20:24:13 GMT
When you measure the bridge rectifier again with the power transistors out of the way you will find that the BR probably measures O.K. Do that first. Bridge rectifier tested with no power transistors: What do you think Frans?
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 6, 2016 20:52:19 GMT
Still O.K. !
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 10, 2016 20:45:26 GMT
back lead on ground: red lead on: C of 2SC5200 (need to measure 1 only) + DC power rail voltage somewhere in the region between +60 and +80V ? C of 2SA1943 (need to measure 1 only) 1 DC power rail voltage somewhere in the region between -60 and -80V ? output of the amp (should read 0V +/- a few mV DC) B of 2SC5200 (need to measure 1 only) expect around +0.8V E of 2SC5200 (need to measure 1 only) expect around +0.1V B of 2SA1943 (need to measure 1 only) expect around -0.8V E of 2SA1943 (need to measure 1 only) expect around -0.1V After our little Skype session Frans I had time to do some measuring. The good news is nothing went BANG C of 2SC5200 = -0
C of 2SA1943 = 0
B of 2SC5200 = -76.2E of 2SC5200 = -76.9B of 2SA1943 = 76.9E of 2SA1943 = 77.3Nothing like what you was expecting i am afraid to say!
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 10, 2016 21:13:57 GMT
Actually I think it is but the test leads were not connected to the right spot.
The C of both parts is the give away here. Instead of measuring from ground I suspect you measured from the power supply line.
In that case 76.9 - 76.9 = 0V will be the emitter (and thus output) voltage 76.9 - 76.2 = +0.7V volt.
On the negative side: 77.3 - 77.3 = 0V on the emitter = output 77.3 - 76.9 = -0.4V (this could be a measurement error)
So you could redo the measurements but with the black testlead connected to the screen of the RCA input. Then measure the voltages on the B, C and E of the transistors (without them installed yet)
They might come up as: B of 2SC5200 (need to measure 1 only) expect around +0.7V E of 2SC5200 (need to measure 1 only) expect around +0.0V C of 2SC5200 (need to measure 1 only) +76.9V B of 2SA1943 (need to measure 1 only) expect around -0.7V E of 2SA1943 (need to measure 1 only) expect around -0.0V C of 2SA1943 (need to measure 1 only) -77.3V
If it does everything will be O.K. and you could fit the 4 transistors that are still O.K. See if it still measures O.K. and test it.
Then replace the transistors with the new ones.
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 11, 2016 21:07:07 GMT
Actually I think it is but the test leads were not connected to the right spot. The C of both parts is the give away here. Instead of measuring from ground I suspect you measured from the power supply line. In that case 76.9 - 76.9 = 0V will be the emitter (and thus output) voltage 76.9 - 76.2 = +0.7V volt. On the negative side: 77.3 - 77.3 = 0V on the emitter = output 77.3 - 76.9 = -0.4V (this could be a measurement error) So you could redo the measurements but with the black testlead connected to the screen of the RCA input. Then measure the voltages on the B, C and E of the transistors (without them installed yet) They might come up as: B of 2SC5200 (need to measure 1 only) expect around +0.7V E of 2SC5200 (need to measure 1 only) expect around +0.0V C of 2SC5200 (need to measure 1 only) +76.9V B of 2SA1943 (need to measure 1 only) expect around -0.7V E of 2SA1943 (need to measure 1 only) expect around -0.0V C of 2SA1943 (need to measure 1 only) -77.3V If it does everything will be O.K. and you could fit the 4 transistors that are still O.K. See if it still measures O.K. and test it. Then replace the transistors with the new ones. New measurments Frans - with the black test lead on the RCA screen: B of 2SC5200 = -77.1E of 2SC5200 = -77.9C of 2SC5200 = - 0B of 2SA1943 = 77.4E of 2SA1943 = 77.9C of 2SA1943 = 0
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 11, 2016 23:00:03 GMT
Sooooo it looks like the output stage is not the most common 'emitter follower' setup but the collectors are the outputs for this amp. Didn't find a schematic anywhere, I am sure it's out there though.
The 0.8V for the 2SC side is a bit high but the 0.5V on the 2SA side is a bit low. That is probably because there are no transistors in there and the amp itself is compensating to get the output to 0V. Which is a positive sign.
0.5 (2SC side) + 0.8V (2SA side) = 1.3V when divided by 2 you get 0.65V per transistor which seems about right.
I would simply solder in 1 2SC5000 and 1 2SA1943 and measure again.
If it doesn't explode connect a speaker (preferably a cheapie) and see if it plays. Just softly, don't crank up the volume.
When it works... just solder in the 1 of each new transistors (after removing the old ones) and see if it functions in the same way. If it does... place the other new transistors and test it shortly (low volume) and then assemble the amp again.
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Jan 17, 2016 16:13:58 GMT
If it doesn't explode connect a speaker (preferably a cheapie) and see if it plays. Just softly, don't crank up the volume. When it works... just solder in the 1 of each new transistors (after removing the old ones) and see if it functions in the same way. If it does... place the other new transistors and test it shortly (low volume) and then assemble the amp again. Hi Frans, I soldered the 4 remaining good transistors back in place and put the amp back together. I then connected it back to my system using a very cheap speaker and to my amazement MUSIC So I am guessing I originally blew one PNP & one NPN transistor in the first place? Now I need to order the replacement transistors but I have a query? You said about populating the TWO SPARE transistor openings and that would give the amp more current, but may get warmer? A: What effect would more current have on the sound and would it be more beneficial when I eventually have them running the Final 03 Electrostatics? B: Do you think it would be a good idea to use the 4 old transistors from this board to populate the other amp board to make that one the same? Thanks Chris
|
|
solderdude
Administrator
measureutternutter
Posts: 4,881
|
Post by solderdude on Jan 17, 2016 20:49:27 GMT
A: I don't think the extra current could be audible. After all it is mostly determined by the power supply. The benefits I do see have to do with spreading short circuit currents over more devices.
This amplifier does not have any over-current protection other than 2 fuses which tend to burn out after the transistors already have given up the ghost. So ... using more transistors in parallel may lower the current in a single transistor when the amp is accidentally shorted or sees a very low impedance load (think electrostats)
The data sheets mention 17A per transistor so with 3 one would be safe to 50A ! Alas the 17A is specified with max 10V across it. When one shorts the amp there is 80V across it though.
I will give an example. Say the output is accidentally shorted and the there are 3 output devices that each 'handle' 1/3 of the total short circuit current with 70V across them. Just 2A with 80V = 160W which is already exceeding the maximum dissipation it can handle (150W) With 3 in parallel a short circuit current of above 6A will result in a blown up amp. With 5 of them this current = 10A. So safer to operate under high output conditions for sure.
The extra 'idle current' the extra transistors will have will make the amp somewhat 'warmer' when 'on' and not playing any music.
B: Good idea !
|
|
Crispy
very active
Madrigal music is playing - Voices can faintly be heard, "Please leave this patient undisturbed."
Posts: 776
|
Post by Crispy on Feb 28, 2016 20:19:07 GMT
A: I don't think the extra current could be audible. After all it is mostly determined by the power supply. The benefits I do see have to do with spreading short circuit currents over more devices. This amplifier does not have any over-current protection other than 2 fuses which tend to burn out after the transistors already have given up the ghost. So ... using more transistors in parallel may lower the current in a single transistor when the amp is accidentally shorted or sees a very low impedance load (think electrostats) The data sheets mention 17A per transistor so with 3 one would be safe to 50A ! Alas the 17A is specified with max 10V across it. When one shorts the amp there is 80V across it though. I will give an example. Say the output is accidentally shorted and the there are 3 output devices that each 'handle' 1/3 of the total short circuit current with 70V across them. Just 2A with 80V = 160W which is already exceeding the maximum dissipation it can handle (150W) With 3 in parallel a short circuit current of above 6A will result in a blown up amp. With 5 of them this current = 10A. So safer to operate under high output conditions for sure. The extra 'idle current' the extra transistors will have will make the amp somewhat 'warmer' when 'on' and not playing any music. B: Good idea ! Sorry for the delay in replying Frans, work and life in general just keeps getting in my way My problem at the moment with the repair is one of the heatsinks. I had to drill and tap new holes into the two heatsinks. The first one went like a dream, the second one a right B*****D. One of the holes tapped out too much? and the bolt would not tighten up the tranistor, so I thought no problem use a self tapping screw. This self tapper snapped off in the hole and it was a right sod trying to re-drill it, then I went to re-tap the hole and guess what the tap broke off in the hole. I was really pissed off with myself and nearly got a big hammer to my amp Anyway after trying to drill the tap out for most of today with an ordinary drill bit I gave in and finally googled how to get it out. It seems I needed a carbide drill bit, so I ordered one and hopefully it will arrive soon. I think once it is drilled out I will insert the new screw into the heatsink with araldite and use nuts to tighten the transistor into place.
|
|