I was looking for a nice small amplifier that would be good for moving around and testing things like preamps and such. I have a Crown D60 which is decent and thin, but I have a friend offer me a Jamo MPA-101. Awhile back he tried to get into electronics, but gave up fairly quickly. He did do a “recap” on it and I’ve gone over the work and it looks decent.
The front LED doesn’t light (it works), it has high offset (200 mV) and it seems to have odd power up behavior. So I wanted to isolate the power supply to make sure it is good to go. Clearly the power on led isn’t working so the power supply isn’t good.
I noticed that there are two large resistors and both are showing signs of heat stress, here is one of them. He is turning brown in the middle.
So I thought I would disconnect everything possible from the power supply board. I removed the fan, the LED, and both amplifier boards. So it is just the one board and the only wires connected to it come from the secondary of the transformer.
I powered up and immediately for items on the board got very hot. There are two 1 watt 15v zener diodes and I would say within 6 to 7 seconds they were at about 230-250 F.
The other two items were the two large resistors which were right at or just slightly above 200 F, the photo is of the cooler of the two.
I’m thinking this might not be normal, not sure if I’ve seen a zener diode get this hot.
The two diodes in question are ZD2 and ZD3. The two resistors are R79 and R80.
The 15 V coming off of the Zener diode is actually closer to 17 V. I understand the voltage drop across those resistors is determined by the current and the heat created is from that current.
Is all this heat strictly being caused by the two zener diodes pulling that voltage to ground? I have not found any shorts on the board, I pulled both of the diodes and they measure perfectly fine out of circuit on a zener diode tester. They are obviously faults on the board as this board is not providing the voltage to the front panel LED, but not sure if these are related. I just want to make sure that the behavior I am seeing with the extreme heat on these two dies and two resistors look OK. If these two resistors are getting this hot strictly from the two diodes pulling the voltage to ground I can imagine they get quite a bit warmer when everything else is connected to the 15 V rails. Well I guess not that much more is pulled. Here’s the other portion of the power supply.
Dan
The front LED doesn’t light (it works), it has high offset (200 mV) and it seems to have odd power up behavior. So I wanted to isolate the power supply to make sure it is good to go. Clearly the power on led isn’t working so the power supply isn’t good.
I noticed that there are two large resistors and both are showing signs of heat stress, here is one of them. He is turning brown in the middle.
So I thought I would disconnect everything possible from the power supply board. I removed the fan, the LED, and both amplifier boards. So it is just the one board and the only wires connected to it come from the secondary of the transformer.
I powered up and immediately for items on the board got very hot. There are two 1 watt 15v zener diodes and I would say within 6 to 7 seconds they were at about 230-250 F.
The other two items were the two large resistors which were right at or just slightly above 200 F, the photo is of the cooler of the two.
I’m thinking this might not be normal, not sure if I’ve seen a zener diode get this hot.
The two diodes in question are ZD2 and ZD3. The two resistors are R79 and R80.
The 15 V coming off of the Zener diode is actually closer to 17 V. I understand the voltage drop across those resistors is determined by the current and the heat created is from that current.
Is all this heat strictly being caused by the two zener diodes pulling that voltage to ground? I have not found any shorts on the board, I pulled both of the diodes and they measure perfectly fine out of circuit on a zener diode tester. They are obviously faults on the board as this board is not providing the voltage to the front panel LED, but not sure if these are related. I just want to make sure that the behavior I am seeing with the extreme heat on these two dies and two resistors look OK. If these two resistors are getting this hot strictly from the two diodes pulling the voltage to ground I can imagine they get quite a bit warmer when everything else is connected to the 15 V rails. Well I guess not that much more is pulled. Here’s the other portion of the power supply.
Dan
Is that temperature as measured by the thermal camera alone or have you used a thermometer/k probe?
That is with the thermal camera alone. I will get a measurement with a thermal probe though.
Edit: well got a measurement. It’s difficult to keep the probe directly on the small diode, but I’m only seeing about 160.
Dan
Last edited:
In that circuit, the less current that is drawn off the supply, the hotter the zeners get. It needs enough over head so that if the load current increases, raw supply drops, or both, the 15V supplies stay in regulation.
It’s interesting that there is such a discrepancy, on this exact exact same power up right after I finished measuring with the pro by took the Flir out (which are known to be accurate) and measured again. Hitting above 250 on the zener.
Is it possible that because the probe is just a hard sphere, touching another hard rounded object that the thermal contact isn’t the best???
Other than the actual temperature, does this seem like normal behavior? These four components getting very hot without really any load on them. I guess the only load being the Zeners pulling to ground.
Dan
Is it possible that because the probe is just a hard sphere, touching another hard rounded object that the thermal contact isn’t the best???
Other than the actual temperature, does this seem like normal behavior? These four components getting very hot without really any load on them. I guess the only load being the Zeners pulling to ground.
Dan
That makes perfect sense. Well, I have to have this board out in order to measure voltages and such, like figuring out why the front light isn’t working. so I will go through and confirm what I can and then put the power supply back in.
Dan
I've had similar discrepancies in temp measurements with a thermal camera showing higher than actual temps. Emmissivity or other settings in the TC perhaps
All true, but if current flows to the load and reduces dissipation in the zener, the series resistors R79 and R80 will get even hotter. Not a good situation. Maybe the transformer output voltage is excessive or the design is lacking?
Can you post more of the schematic?
Good luck.
Can you post more of the schematic?
Good luck.
Absolutely. Try some thermal paste (PC CPU cooler thermal paste, if non-metallic). But I'd use the FLIR anyway, way faster and easier 🙂
For a more direct attack on the issue, I suggest measuring the DC voltage across the 180R resistors. Then compute power dissipated, =V*V/180. How does computed power comport with the Wattage of the installed resistors?
I just finished going through both amplifier boards, checking all diode, checking for any shortage, transistors, and checking resistors. I have read that this model also suffers from solder joints as the wave soldering wasn’t the best and heat they tend to crack. So every joint was flowed with solder.
I fully reassembled the amplifier and went for it. I have found your statement to be the case.
Resistors are getting warm, but R79 (circled) is getting to a whole new level of hot nearing 450. No wonder it’s turning brown. The associated zener ZD2 doesn’t look like it’s really putting off any heat at all.
Here is the other side of the filter, cap, there you will find R80 which is still getting warm, but nowhere near the same temp as R79. ZD3 is the only one putting off heat. Before when I had the power supply board out, you saw two dots they’re emitting heat, now just the one.
So the zener that is stain cool is cooking its resistor. I measured the voltage of the +15v rail on one of the amplifier boards and it’s at 10.8v. So it’s been dragged down pretty good.
I will post the whole schematic, see how this does. If it doesn’t look good, then I’ll have to post it in sections
Dan
I am most certainly having an issue and it’s clear the resistor is dissipating more wattage than it should be. I’m going to be placing an order for 82 ohm resistors in 2w up to 5w. Once I have it running correctly and I get the voltage drop. I will replace both of those resistors so that they are the same. R79 is too far damaged and I’m going to replace it regardless even though the resistance is OK. It seems worse now than it did this morning, in looks. The originals look to be 2 watt. I will likely fit a three or four if they physically fit.
Edit: you’d think if something was pulling the 15v down it would be getting hot right?
Dan
If you're referring to the 15V zener, if the load network is drawing too much current and pulling the 15V rail below the nominal 15V, current through the 15V zener will drop dramatically and dissipation drops accordingly. The zener is acting as a shunt regulator, and starts to conduct as the applied voltage approaches 15V. It clamps the voltage from rising above 15V.
I can't find the +Vcc and -Vcc voltages noted. What do you measure?
Yes, that is exactly how it is behaving. Since the voltage is being pulled down that zener is obviously running cool. Yeah, I thought that was kind of stupid. They didn’t provide rail voltages. I’m seeing a +29.4v and -30v. I’m guessing it’s also pulling the main rail down a little bit as well.
Dan
So expected dissipation in each 180R would be about 1.25W. Seems like there's probably a defect in the +/- 15V circuitry causing excessive current draw.
Nominal current drawn from the +/- 15V rails should be less than 83mA.
Nominal current drawn from the +/- 15V rails should be less than 83mA.
Yes, I would say there are definitely issues. My voltages look like they are all over the place. I think I am going to remove the power supply board again and measure the voltages with no load on them. See if my + and - 15v is present. It sucks though, I have to desolder a bunch of header pins in order to remove the board. I gave false information above, I was measuring at ground from the wrong side.
Each amplifier board has a pin header, where the +/- 15v and + 12 is present. One channel uses 8 pins and the other channel uses 12.
Here is the first channel the number on the left-hand side is what is supposed to be there according to the schematic. The number on the right hand side is what is actually being measured. so on the photo below the 12 V looks pretty good, the - 15 is a bit high and the + 15 is way out.
The other channel doesn’t look any better. On this channel the +12 is gone, and not one single voltage looks correct.
Dan
The amplifier board that has the 12 pin header also has another pin header that supplies voltages too the input board. I wanted to disconnect that and measure again to see if that is altering voltages at all.
It changed the voltages on that amplifier board a good deal. The new ratings are marked in red. Pin 6 now has voltage, the wrong polarity. I don’t understand how pins 5 through 7 are all the wrong polarity.
Here is the other channel, not really affected. My next step will be to remove the power supply board to measure the voltages on those headers without the amp boards hooked up. That will likely be tomorrow as I am on childcare duty.
Dan
It changed the voltages on that amplifier board a good deal. The new ratings are marked in red. Pin 6 now has voltage, the wrong polarity. I don’t understand how pins 5 through 7 are all the wrong polarity.
Here is the other channel, not really affected. My next step will be to remove the power supply board to measure the voltages on those headers without the amp boards hooked up. That will likely be tomorrow as I am on childcare duty.
Dan
Well, I managed to escape with just enough time to remove the power supply and measure voltages. I have no idea why on earth I didn’t think to do this earlier. I alway check for voltages, but I was just too focused on the main rails I guess.
It looks like both pin headers share a + and - 15v rail and the positive is absent. So that right there tells me there is an issue on the power supply board. the larger pin header has an additional separate rail, and both of those are present. Also the 12 V is present on both as well. I think when I’m going to do is repair the 15 V rail on the power supply so that all are present and accounted for. I will then hook up one channel at a time, starting with the channel that only has one set of 15 V rails.
It looks like this is the 15 volts that is missing so when I get back downstairs, the first thing I will check is Q15.
Dan
It looks like both pin headers share a + and - 15v rail and the positive is absent. So that right there tells me there is an issue on the power supply board. the larger pin header has an additional separate rail, and both of those are present. Also the 12 V is present on both as well. I think when I’m going to do is repair the 15 V rail on the power supply so that all are present and accounted for. I will then hook up one channel at a time, starting with the channel that only has one set of 15 V rails.
It looks like this is the 15 volts that is missing so when I get back downstairs, the first thing I will check is Q15.
Dan
Last edited:
lol, and of course Q15 is measuring completely fine on this Peak DCA75.
So there must be an issue on the base side of Q15. I did check R50 and R52. Obviously the 15 V is present on the emitter, but something is preventing it from being passed to the collector. Guess I need to check Q8, D2, and a couple of other things. I would have checked D2 in circuit.
Edit: Hmmmm, Q8 and D2 both measure good out of circuit. R49, R81, and R53 all measure dead on. Thoughts? Is it possible that one of the two transistors measure fine out of circuit but once they see actual voltage and current, they fall apart???
Dan
So there must be an issue on the base side of Q15. I did check R50 and R52. Obviously the 15 V is present on the emitter, but something is preventing it from being passed to the collector. Guess I need to check Q8, D2, and a couple of other things. I would have checked D2 in circuit.
Edit: Hmmmm, Q8 and D2 both measure good out of circuit. R49, R81, and R53 all measure dead on. Thoughts? Is it possible that one of the two transistors measure fine out of circuit but once they see actual voltage and current, they fall apart???
Dan
Last edited: