Hello.
I recently picked up a PG Tantrum 500.2 which was reported to be dead. On checking the output FETs all six in the right channel were short so I've replaced them. Once I bolted everything back together and powered it up without an input and no load connected the amp started to draw in excess of 10A so I removed the power immediately. The only way I can power up the amp is to fully wind down the bias on the right channel. When this is done the amp will draw 4A again without an input and no load connected. All the output FETs in the right channel start to get warm were the left channel is cold. I've measured the voltage at source on the output FETs in the left side and got 47mV. When measuring the source on the output FETs on the right channel and try to wind up the bias I get to around 30mV and the amp is drawing 10A plus again and again the FETs are getting warm.
I'm now at a bit of a loss as to where to look next. I have limited knowledge of electronics and it's been a while since I've worked on anything like this so any help would be much appreciated.
Thanks
Nick
I recently picked up a PG Tantrum 500.2 which was reported to be dead. On checking the output FETs all six in the right channel were short so I've replaced them. Once I bolted everything back together and powered it up without an input and no load connected the amp started to draw in excess of 10A so I removed the power immediately. The only way I can power up the amp is to fully wind down the bias on the right channel. When this is done the amp will draw 4A again without an input and no load connected. All the output FETs in the right channel start to get warm were the left channel is cold. I've measured the voltage at source on the output FETs in the left side and got 47mV. When measuring the source on the output FETs on the right channel and try to wind up the bias I get to around 30mV and the amp is drawing 10A plus again and again the FETs are getting warm.
I'm now at a bit of a loss as to where to look next. I have limited knowledge of electronics and it's been a while since I've worked on anything like this so any help would be much appreciated.
Thanks
Nick
Hi Nick,
They use a unique design where the bias is actually controlled at the very front end of each main amp from a Op-amps voltage reference located on the 12 volt side of the power supply.
With 47 MVDC output at idle no signal says somethings leaking very badly in the channel and since you have replaced all the outputs its is most definitely not them. gate resistors get strained on failures like you describe and also many of the other devices all the way back to the very input pair. there are also several zener diodes that can be damaged out of tolerance.
Gate drivers in this amp MPAS06 and MPSA56 transistors can become damaged by this sort of failure so I would find a bunch of them and start replacing them. they cost me about 3 cents each so its simple math to just replace them. There are 4 each 6.0 volt and 2 each 12.0 volt zener diodes in each channel and they like to leak badly on a failure like this so here again the small price for new ones dictates replacement. All the rest of the diodes are 1n4148 and are even cheaper to replace. Oh and there is another 6.0 zener located near the bias pot on the left channel.
The symptoms you describe on retest of the amp is what is telling me to tell you to start replacing the devices listed above. Also check all resistor for out of tolerance condition, and bad solders on all the large wattage value resistor, and the plastic power transistors standing up right in free air. I have seen these amps get pretty much melted by owners.
All the little resistors are 1/8 watt 1% tolerance so they should read almost exactly what the color codes say they should be.
Also mark the original bias set points on the pots with a marker for reference only and turn them both back to zero bias while testing and working on the amp. Once your clear of all issues you can revisit these set points and set them correctly to spec after all repairs are completed.
It costs me about $6.00 and change to completely replace the entire channel your costs will be higher depending on how hard you shop for parts. Oh 10 MVDC or less is what i consider acceptable on the outputs of this amp once its repaired, and that is at any bias or input gain level.
And if your in question about any parts value just post about it giving its part number as seen on the PC board. I'll help you out ...C
They use a unique design where the bias is actually controlled at the very front end of each main amp from a Op-amps voltage reference located on the 12 volt side of the power supply.
With 47 MVDC output at idle no signal says somethings leaking very badly in the channel and since you have replaced all the outputs its is most definitely not them. gate resistors get strained on failures like you describe and also many of the other devices all the way back to the very input pair. there are also several zener diodes that can be damaged out of tolerance.
Gate drivers in this amp MPAS06 and MPSA56 transistors can become damaged by this sort of failure so I would find a bunch of them and start replacing them. they cost me about 3 cents each so its simple math to just replace them. There are 4 each 6.0 volt and 2 each 12.0 volt zener diodes in each channel and they like to leak badly on a failure like this so here again the small price for new ones dictates replacement. All the rest of the diodes are 1n4148 and are even cheaper to replace. Oh and there is another 6.0 zener located near the bias pot on the left channel.
The symptoms you describe on retest of the amp is what is telling me to tell you to start replacing the devices listed above. Also check all resistor for out of tolerance condition, and bad solders on all the large wattage value resistor, and the plastic power transistors standing up right in free air. I have seen these amps get pretty much melted by owners.
All the little resistors are 1/8 watt 1% tolerance so they should read almost exactly what the color codes say they should be.
Also mark the original bias set points on the pots with a marker for reference only and turn them both back to zero bias while testing and working on the amp. Once your clear of all issues you can revisit these set points and set them correctly to spec after all repairs are completed.
It costs me about $6.00 and change to completely replace the entire channel your costs will be higher depending on how hard you shop for parts. Oh 10 MVDC or less is what i consider acceptable on the outputs of this amp once its repaired, and that is at any bias or input gain level.
And if your in question about any parts value just post about it giving its part number as seen on the PC board. I'll help you out ...C
Last edited:
Hi C,
Have you seen any problems with the 10K thermistors in these amps? Only reason I ask is that I was checking over a few things this evening and noticed that on initial power on the amp draws around 1.5A and over a period of 15-20 minutes this increases to 4A. Was just thinking that if there was a problem with it and as it warms up the voltage drop across R220, R221, R222, R223 will change and thus effect how the transistors are biased. Would I be right in thinking this or should I still be going down the line of replacing everything?
Thanks
Nick
Have you seen any problems with the 10K thermistors in these amps? Only reason I ask is that I was checking over a few things this evening and noticed that on initial power on the amp draws around 1.5A and over a period of 15-20 minutes this increases to 4A. Was just thinking that if there was a problem with it and as it warms up the voltage drop across R220, R221, R222, R223 will change and thus effect how the transistors are biased. Would I be right in thinking this or should I still be going down the line of replacing everything?
Thanks
Nick
Afternoon Nick,
Are your bias pots zero'ed like mentioned in my first post?
Remember what i said in my first post this amp is biased differently at its input pairs. And its output temp sense tracking transistor although sink mounted connects back at the input pair.
The Bias temp sense transistor is Q200 and its on the sink near the outputs, and its directly connected back to the +37 volts supply which is there just for this purpose, and its wired back to the input pairs like a temp sense current source would be utilized.
Q 204 has a base signal from a 339 and Q10 located on the 12 volt side as a reference and the bias pot is connected to this transistors emitter and it is supplied -34 VDC from the amp rail, and this transistor connects to the input pair on their emitter side of Q201 and Q207.
Left channel has a 6.0 volt zener, a 100K resistor and a 1ufd/50 cap across the base and the -34 VDC supply only on the left channel so if you mirror this AB don't let the extra devices on the left channel throw you, they are not common to both channels but I don't think they need to be to achieve what PG had in mind.
The Bias signal from the supply is labelled "Decoupled 5.1 volt reference", and it also has twin temp sensors for its control.
R222 is a temp sensor but its not connected to the sink so its for ambient temp sensing. Can it affect bias ? Yes but its only supposed to adjust for ambient temps not output temps. < not as much control as the other sensor>
Can it be damaged ? Possible but so would the other three resistors that connect to it to form a bridge, so if none of the other resistors are damaged them I am doubtful the temp sensor got scotched. Simple high and low temps applied and resistance readings will tell.
The symptoms your telling me about I have seen a few dozen times so I advise you as I would myself and remove all questionable devices and replace with new that were directly connected to the shorted fets which folded the rails back onto them, and likely caused excessive heat damage to the silicon structures. I have had to replace even the input pairs on these just to get "Like New" performance of the amp.
This amp bias's its input pairs and temp sense back to the input pairs, not the drivers prior to the outputs like most other amps you see everyday.
tech note: Channel left is 1XX part code and channel right is 2XX part codes
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.