I'm Having a problem with a Taramps HD 15000.
I’ve desoldered the SD pad so it cuts the 5V feed to the MCU. The amp comes out of protect. With the new fet drivers installed they will not oscillate on the fet gates. The PWM input on pin 1 of the driver is getting a nice clean square wave. The powers and grounds check out ok.
What is weird is the drivers are getting 5V on the SD pins which is causing them to shutdown and not oscillate. What’s really more weird is they are outputting 10V on pins 14 - Driver A GND and pin 15 - driver A output (high side). Is this because the fets are out? Does this amp need the fets to oscillate?
This amp also draws 2A at idle as well. I’ve got my power supply on 14V with no more than 3A allowable current. Also as of now only the fets are out as I don’t want them to go full blast and blow up with 10V on their gates.
I’ve replaced a couple damaged zener diodes and both fet drive ic’s (to be safe). Is the MCU supposed to pull the 5V down to ground and that’s why nothing will oscillate when the SD pad is soldered?
I’ve desoldered the SD pad so it cuts the 5V feed to the MCU. The amp comes out of protect. With the new fet drivers installed they will not oscillate on the fet gates. The PWM input on pin 1 of the driver is getting a nice clean square wave. The powers and grounds check out ok.
What is weird is the drivers are getting 5V on the SD pins which is causing them to shutdown and not oscillate. What’s really more weird is they are outputting 10V on pins 14 - Driver A GND and pin 15 - driver A output (high side). Is this because the fets are out? Does this amp need the fets to oscillate?
This amp also draws 2A at idle as well. I’ve got my power supply on 14V with no more than 3A allowable current. Also as of now only the fets are out as I don’t want them to go full blast and blow up with 10V on their gates.
I’ve replaced a couple damaged zener diodes and both fet drive ic’s (to be safe). Is the MCU supposed to pull the 5V down to ground and that’s why nothing will oscillate when the SD pad is soldered?
Last edited:
After doing more reading, I’ve mounted the output fets and resoldered the SD check pad. I power up the amp and it comes alive and oscillates. Note to self these amps must have all the fets to switch ground.
When checking for heat, the side that was blown is warm but the side that was not blown is much warmer if not hot to the touch after about a minute. I’ve checked the gate resistors and they are all ok. I’ve got a good square wave on my high side and low side. My offset at the speaker terminals was around 0.320V. I don’t see an adjustment pin so I adjusted it using the offset pot and got it down to around 0.10V. I know that probably won’t work since the amp should idle for about 2 minutes first but I feel like the outputs are getting too hot on one side.
What could be causing this extra heat? Should I just mount this baby to the heat sink and go from there?
When checking for heat, the side that was blown is warm but the side that was not blown is much warmer if not hot to the touch after about a minute. I’ve checked the gate resistors and they are all ok. I’ve got a good square wave on my high side and low side. My offset at the speaker terminals was around 0.320V. I don’t see an adjustment pin so I adjusted it using the offset pot and got it down to around 0.10V. I know that probably won’t work since the amp should idle for about 2 minutes first but I feel like the outputs are getting too hot on one side.
What could be causing this extra heat? Should I just mount this baby to the heat sink and go from there?
I've been checking this board now for a little bit and here is what I've found.
On the side that doesn't heat up, there's a DZ3 diode that is between the high and low sides. There is 12V across that diode. On the side that heats up, DZ1 has 10.5V. Also on the side that doesn't heat up, all gate zener diodes have 5.8V across them. On the side that heats up the high side has 5.1V across the gate zener and the low side has 5.8V across the zener.
What should I be checking? I have removed all diodes on the heated side and checked them individually and they're checking OK. I don't have a diagram and the only one that I'm using that resembles closely is the HD 3000.
On the side that doesn't heat up, there's a DZ3 diode that is between the high and low sides. There is 12V across that diode. On the side that heats up, DZ1 has 10.5V. Also on the side that doesn't heat up, all gate zener diodes have 5.8V across them. On the side that heats up the high side has 5.1V across the gate zener and the low side has 5.8V across the zener.
What should I be checking? I have removed all diodes on the heated side and checked them individually and they're checking OK. I don't have a diagram and the only one that I'm using that resembles closely is the HD 3000.
I've got one that's acting in similar manner that I haven't messed with too much yet so please keep posted on here what's going on. Do you happen to have the schematic. Taramp tech support told me to kick rocks pretty much. I'm waiting on my new oscilloscope to come in the mail before I get back to work on mine. My old HP scope has gotten about useless.
I’ve contacted Taramps for schematics and they will not respond. I’ve contacted all 3 warranty centres they list and one got back to me and said they are not allowed to sell me schematics or anything else.
I’m scoping mine right now and I’m finding some weird things. For starters there’s a difference between the gate voltages between both high sides. I’m troubleshooting that now. I’m assuming that the supply voltage on the driver chip at pin 16 is what gets fed to the gate. I switched the drivers around and same result - fets start heating up and I’ve gotta shut it down to cool off.
Only thing left is there has to be something pulling the voltage down. Without a schematic it’s a PITA.
I will keep everyone updated as to what is found and what exactly I had to do to solve it. I love learning this stuff. Without Perry’s electronics course on this stuff I wouldn’t have made it as far as I have.
I’m scoping mine right now and I’m finding some weird things. For starters there’s a difference between the gate voltages between both high sides. I’m troubleshooting that now. I’m assuming that the supply voltage on the driver chip at pin 16 is what gets fed to the gate. I switched the drivers around and same result - fets start heating up and I’ve gotta shut it down to cool off.
Only thing left is there has to be something pulling the voltage down. Without a schematic it’s a PITA.
I will keep everyone updated as to what is found and what exactly I had to do to solve it. I love learning this stuff. Without Perry’s electronics course on this stuff I wouldn’t have made it as far as I have.
What I have gathered tonight is:
On the side that heats up, I have checked the driver chip for all voltages and compared with the side that does not heat up. All voltages are the same with the exception of the following:
Pin 15 - Should be 12V, only 10.5V
Pin 16 - Should be 12V, only 10.5V
When I trace it with the scope with two channels lined up together, they do not match as the rise isn't the same. It's off only a little bit. It seems as if the driver chip is not getting enough voltage to drive the gate of the high side fets.
I've checked the DR1 diode right in front of it and it's ok. I've even switched positions of this diode with the side that's not heating up and there's no change. I'm now assuming there's something eating up 2V or so of my available voltage and it baffles me as to where to look.
On the side that heats up, I have checked the driver chip for all voltages and compared with the side that does not heat up. All voltages are the same with the exception of the following:
Pin 15 - Should be 12V, only 10.5V
Pin 16 - Should be 12V, only 10.5V
When I trace it with the scope with two channels lined up together, they do not match as the rise isn't the same. It's off only a little bit. It seems as if the driver chip is not getting enough voltage to drive the gate of the high side fets.
I've checked the DR1 diode right in front of it and it's ok. I've even switched positions of this diode with the side that's not heating up and there's no change. I'm now assuming there's something eating up 2V or so of my available voltage and it baffles me as to where to look.
Thank you for the updates, let me know if you need me take any measurements on mine for you to compare to. Late tomorrow I'll have my new scope. I agree about the schematics would be very helpful.
I sent a message to Taramps tech support page last Saturday. I got a reply today from a Taramps amp repair tech.
He said: (For the HD 15K)
- Amp with all fets in @14.4V should idle around 5.1A. It might fluctuate from around 4.5A to 5.1A but thats normal.
- The negative output section will warm up faster than the positive and that is normal.
- The voltage for DC offset must be measured at the speaker terminals and should be below 1.0V. If not then adjust it (I thought that 1.0V might be high but OK)
- When the amplifier's red protect light flashes twice in a row with a pause it usually means you just met the under current condition.
- He also said that my measuring voltages the way I was is OK but you need to be measuring the whole circuit's draw at different input voltages to verify amp integrity
- If there is a weird problem then the amp draw would go out of range (too high or too low) and finding the problem would be easier to detect
- During amp idle, the A.P.C terminals should be 2.7V.
- The Feed pad on the mainboard, when desoldered separates the audio return signal from the outputs back to the audio drive circuit.
What I learned so far about Taramps HD 15K boards:
- Desolder the SD pad on the mainboard to trick the MCU into coming out of protect. This is basically to isolate the MCU protection from the output driver chip protection. The SD pins on the driver IC's and the MCU always have a constant 5.0V on them causing them to shut down. When the outputs are in, and the amp is working, that 5V on the SD terminals gets pulled down to 0V through a 1K resistor that feeds both the driver IC's.
- Check with a magnifying glass all solder joints - yes this can get timely, but it will save you. I've found too many solder joints right from factory that had not been fully soldered. The amp I'm working on is not very old.
- All drivers and fets need to be working in order to come out of protect properly
- The fets I removed are IRGP6690D. These are no longer available so I got what they come with now - IRGP4066D-EPBF. This seems to be a direct replacement. I did not have to modify gate resistors. The gete charge and gate capacitance are quite close on these.
- The zener diodes that I replaced are MELF 15V 1W - CLL4744ACT
My board doesn't have any chips that have their part number grounded off, which makes it easier to work through it. Tomorrow will tell the tale as I will power it up with 100A to start and I'll feed an audio signal into it.
He said: (For the HD 15K)
- Amp with all fets in @14.4V should idle around 5.1A. It might fluctuate from around 4.5A to 5.1A but thats normal.
- The negative output section will warm up faster than the positive and that is normal.
- The voltage for DC offset must be measured at the speaker terminals and should be below 1.0V. If not then adjust it (I thought that 1.0V might be high but OK)
- When the amplifier's red protect light flashes twice in a row with a pause it usually means you just met the under current condition.
- He also said that my measuring voltages the way I was is OK but you need to be measuring the whole circuit's draw at different input voltages to verify amp integrity
- If there is a weird problem then the amp draw would go out of range (too high or too low) and finding the problem would be easier to detect
- During amp idle, the A.P.C terminals should be 2.7V.
- The Feed pad on the mainboard, when desoldered separates the audio return signal from the outputs back to the audio drive circuit.
What I learned so far about Taramps HD 15K boards:
- Desolder the SD pad on the mainboard to trick the MCU into coming out of protect. This is basically to isolate the MCU protection from the output driver chip protection. The SD pins on the driver IC's and the MCU always have a constant 5.0V on them causing them to shut down. When the outputs are in, and the amp is working, that 5V on the SD terminals gets pulled down to 0V through a 1K resistor that feeds both the driver IC's.
- Check with a magnifying glass all solder joints - yes this can get timely, but it will save you. I've found too many solder joints right from factory that had not been fully soldered. The amp I'm working on is not very old.
- All drivers and fets need to be working in order to come out of protect properly
- The fets I removed are IRGP6690D. These are no longer available so I got what they come with now - IRGP4066D-EPBF. This seems to be a direct replacement. I did not have to modify gate resistors. The gete charge and gate capacitance are quite close on these.
- The zener diodes that I replaced are MELF 15V 1W - CLL4744ACT
My board doesn't have any chips that have their part number grounded off, which makes it easier to work through it. Tomorrow will tell the tale as I will power it up with 100A to start and I'll feed an audio signal into it.
Last edited:
- Status
- Not open for further replies.