Problems with an amp

Hello
I have a 2 channel Audiobahn A/B class amplifier (A8002T), I bought it brand new in the US, when I came to my country (Mexico) I connected it and it was working fine until I turned the volume up and there was a short circuit in the system, I checked the fuses and they were blown, I replaced them with new ones and turned it on again, and it happened again, but this time the MOSFET IRFZ44N were blown, This time I replaced them because I tought they were defective and again, the new ones were blown...

I plan to get a new MOSFETS but with a bigger value, like a 70A drain.

Do you think this could be a fix?
The warranty was only valid in the US, that's why I couldn't make it valid.

Thank you
 
Re: Problems with amp

stevenbrenner said:
I would check the impedance of the speaker system so as not to have this happen again.

Yes, I checked the impedance of the sub, nothing wrong with it 4 ohms, that is what the amp says it supports.

richie00boy said:
Replacing with higher current MOSFETs is a bad idea. They have blown again because there is a fault, most probably them blowing in the first place has caused more damage than just the MOSFETs.


I guess I'll just trow it in the trash can, I shouldn't have bought Audiobahn in the first place.

Thank you all
 
When the MOSFETs die, there ususally isn't much other damage. If there was, it'd be restricted to the totem pole bipolar transistors that drive the MOSFET gates, or the passive components surrounding the MOSFETS.

When you have the MOSFETs short, the amp will blow the fuse as soon as power is applied, regardles of whether the remote is connected, or if it has signal and speakers connected.

WHen you repalce the MOSFETS, id'd just stick in the usual IRF-z44's. Or something with equal or higher ratings. Be sure to replace all of the PSU MOSFETS when you do,

AudiBahn is gnerally considered a middle of the road company. Their amps shouldn't explode. Your sub is the correct rating for the amp. Yo might want to check the sub wiring for shorts to itself or the car chassis. You might also want to check if the amp is overheating - if it runs hot, components are stressed very hard - hook up a computer fan from remote to ground. You should also make sure that the input wiring if of sufficient guage. Sometimes if you have a power power source for the amp and the voltage sags too much, the PSU will try to draw the same wattage which means it will draw more curent and fry (popular failure method of SONY XPLODs). You can measure the voltage at the amp terminals with an ANALOG meter to check for voltage drop while it is playing. The solution is to increase the wire, make sure the connections are good, and maybe get a newer car battery. Farad caps are band-aids. DO you routinely drive the amplifier into clipping? Does the protection on the amp ever engage? It's my goal to install a system so that the protection never activates.
 
Dr. Photon said:
When the MOSFETs die, there ususally isn't much other damage. If there was, it'd be restricted to the totem pole bipolar transistors that drive the MOSFET gates, or the passive components surrounding the MOSFETS.

When you have the MOSFETs short, the amp will blow the fuse as soon as power is applied, regardles of whether the remote is connected, or if it has signal and speakers connected.

WHen you repalce the MOSFETS, id'd just stick in the usual IRF-z44's. Or something with equal or higher ratings. Be sure to replace all of the PSU MOSFETS when you do,

AudiBahn is gnerally considered a middle of the road company. Their amps shouldn't explode. Your sub is the correct rating for the amp. Yo might want to check the sub wiring for shorts to itself or the car chassis. You might also want to check if the amp is overheating - if it runs hot, components are stressed very hard - hook up a computer fan from remote to ground. You should also make sure that the input wiring if of sufficient guage. Sometimes if you have a power power source for the amp and the voltage sags too much, the PSU will try to draw the same wattage which means it will draw more curent and fry (popular failure method of SONY XPLODs). You can measure the voltage at the amp terminals with an ANALOG meter to check for voltage drop while it is playing. The solution is to increase the wire, make sure the connections are good, and maybe get a newer car battery. Farad caps are band-aids. DO you routinely drive the amplifier into clipping? Does the protection on the amp ever engage? It's my goal to install a system so that the protection never activates.

Thanks for this good reply, I will check all of those things you mention, I guess I'm running an 4 gauge wire to the battery and this could be a problem, I'll replace it with an 8 gauge size, I'll also check all the connections, and will check for voltage drop, I hope all this work.

Thank you very much!
 
4 ga should be more than fine. The only weak link in the power deivery system might be a weak alternator or battery (dimming headlights?). The actual amplifier part of the amp should be fine i think.
I would just replace *all* MOSFETS, check to be sure the rest of the parts around them are OK, and then see if it works fine after that.
 
jarolda said:


Thanks for this good reply, I will check all of those things you mention, I guess I'm running an 4 gauge wire to the battery and this could be a problem, I'll replace it with an 8 gauge size, I'll also check all the connections, and will check for voltage drop, I hope all this work.

Thank you very much!

Ooopps, I mean I'm running 8 awg size, will replace with 4 awg size.
 
the amp is fused at 60 amps, and for that you should use something a bit bigger than 8 guage (right on the limit, or slightly past). I recommend 4 gauge. If it doesn't "fix" the amp-blowing-up problem, it should still help your performance, and allow you to later add a distribution block and add more amps. Just be sure that you fuse at the battery. You can use either an AGU fuse (the big tubular glass kind) or an ANL (the flat screw down waffer fuses, available in many crazy watages). There are also Mini-ANL (MTX/Street wires proprietary???) and MAXI fuses (like really big regular car fuses), but they seem to be harder to find. The AGU is a good choice if you aren't planning on adding a second amp. they seem to be commonly available in 40-80 amp sizes. same for MAXI. ANL's go from around 100 to 300 in incremens of 50 amps. mini-ANL seems to be in the middle. Usually if you get a wiring "kit" with 10 or 8 ga, you usually get an integrated ANO (regular car fuse, good to 30 amps, 40s exist but aren't popular) fuse, or sometimes an AGU. 4 gauge kits almost always come with AGUs.

example:
My friend installed a single "340" watt amps and sub in his truck, with a 10 ga kit. The amp and the kit were both fused with 30 amo ATO fuses. Later he gets a Jensen "740" watt jensen fused at 50 (2 x 25 ATO's) to power a second sub. We also got some 4 ga to replace the 10 ga. Then there was an AGU fuse holder with an 80 on the battery, going into a 4 way fused distribution block. the D-block has three inputs that take 2-4 ga, and 4 slots for AGU fuses that can accept 8 ga wire. We used a 40 amp AGU fuse in the D-block to protect the Dual amp and it's recycled 10 ga feed. The Jensen was connected directly to annother input terminal with more 4 ga wire (no need to use a fuse if the guage is not reduced). Later, I gave him my old 1800 watt power inverter, whice draws 150 amps and is fused at 180 (SIX 30 amp ATO's). This was also connected to the distribution block's input with more 4 guage. We also replaced the AGU fuse holder at th battery with an ANL fuse hlder with a 200 amp fuse. The good thing about using a heaveri than needed gauge wire is that you get minimal voltage drop, and you don't need to run more wire later when you want to add more toys :)
 

nickch

Member
2005-07-05 6:50 am
Chch
i had the same problem before, when my 2* IRFZ44 kept blowing... with significant damage on the source pin.
all i did was replace all the resistors connected to those MOSFETs and the MOSFET itself with values taken from the matching pair of MOSFETS next to them. problem was resolved after that...
 
nickch said:
i had the same problem before, when my 2* IRFZ44 kept blowing... with significant damage on the source pin.
all i did was replace all the resistors connected to those MOSFETs and the MOSFET itself with values taken from the matching pair of MOSFETS next to them. problem was resolved after that...


Oh, that sounds interesting, did you replace the resistors with ones with the same value?

Thank you
 

nickch

Member
2005-07-05 6:50 am
Chch
yeah.. they were 4 FETs in total.. but only 2 of them them blowing.. and they were actually 5n60s... what i did was i replaced all 4 with IRFZ44s which were similar to 5n60s and replaced the gate,drain and source resistors on the blown FETs with new ones with equal values as the other 2 unaffected FETs...

another member on this forum suggested that it might have been the PWM chip.. so i replaced that as well.. since it only costed me about 2 NZD here...

had no problems after that..

the weird thing about my problem is that.. once the FETs blow.. the amp would still be able to output music...
another source told me that i prolly blew the supply to either the positive or negative phase or something like that and that the cones on my speaker would only move one-way...
 
nickch said:
yeah.. they were 4 FETs in total.. but only 2 of them them blowing.. and they were actually 5n60s... what i did was i replaced all 4 with IRFZ44s which were similar to 5n60s and replaced the gate,drain and source resistors on the blown FETs with new ones with equal values as the other 2 unaffected FETs...

another member on this forum suggested that it might have been the PWM chip.. so i replaced that as well.. since it only costed me about 2 NZD here...

had no problems after that..

the weird thing about my problem is that.. once the FETs blow.. the amp would still be able to output music...
another source told me that i prolly blew the supply to either the positive or negative phase or something like that and that the cones on my speaker would only move one-way...

Thank you, I'll give it a try, you'll hear from me soon, I hope it solves my problem!

Thank you again
 
nickch said:
yeah.. they were 4 FETs in total.. but only 2 of them them blowing.. and they were actually 5n60s... what i did was i replaced all 4 with IRFZ44s which were similar to 5n60s and replaced the gate,drain and source resistors on the blown FETs with new ones with equal values as the other 2 unaffected FETs...

another member on this forum suggested that it might have been the PWM chip.. so i replaced that as well.. since it only costed me about 2 NZD here...

had no problems after that..

the weird thing about my problem is that.. once the FETs blow.. the amp would still be able to output music...
another source told me that i prolly blew the supply to either the positive or negative phase or something like that and that the cones on my speaker would only move one-way...

Every car amp I have seen up to this point has an even number of power supply FETs. The main transformer has two windings, which are opposite to each other, and the FETs are organized in two banks which alternate driving the transformer. This is sort of a “push pull” way to generate AC flux in the core.

If any whole bank of FETs is gone, the amplifier will still operate, but its power supply will be less efficient, and it will develop slightly less rail voltage, and therefore less output voltage, and will have reduced current capacity. There should be no issue with the speakers only moving in one direction.

An amp will continue to function with only a single FET in the power supply needed. But, this would mean a major loss in current, if the amp intended to have many more.

If all else fails and you don’t know what is up, you could resort to removing all the power supply FETs and sticking a single known good one in there, then trying it. Ideally you should start off with two known good ones, with one driving each half of the transformer.
 
problems with an amp

Hi Jarolda,

You've received a lot of good responses but the following may be helpful (and possibly long and boring) in repairing the amplifier.

If the power supply continues to blow FETs, the problem 'could' be in the power supply but it may be in the audio section. You need to check the audio output transistors. If any output transistor is shorted between pins 2 and 3, the transistor is defective. If the protection circuits in the amplifier aren't working properly (or don't exist), the shorted output transistor would cause the audio section to draw a significant amount of current. This could blow the power supply even if you've replaced all of the defective components. Any defective output transistors must be replaced. If only one output transistor of a bank of parallel transistors is blown, you may remove only the defective transistor to continue with the testing. When you get the amp to a point where it powers up and plays, all parallel transistors (in parallel with the defective one) will have to be replaced.

If the audio output transistors are not shorted, remove all of the FETs from the power supply and connect the amp to the power supply via a relatively small fuse (10 amp should be fine). With the amplifier powered up (including remote) measure the DC voltage on pins 9 and 10 of the power supply IC. They are the last two pins on the top side of the chip on the right side (assuming that the chip is a TL594 or some variant). You should see approximately 1/2 of the B+ input voltage. If you don't, you need to measure the DC voltage on all of the pins and then post them. Look up the datasheet for the chip so you can get the correct number for each pin.

Note: Using a multimeter on 'DC volts' to determine whether the IC is running (producing pulses) is not 100% accurate. Since it's unlikely that you have an oscilloscope, if your meter has a frequency counter, check the frequency on 9 and 10. If you read something near 25khz, you know for sure that the IC is producing pulses.

If the power supply IC seems to be producing output, you then need to check the voltage on the gate pad (where pin 1 of the FET solders into the board). With the black meter lead on ground, the voltage should be approximately the same as you read on the chip. Do this on ALL of the gate pads. If they have ~1/2 voltage on them, then it's likely that the driver transistors between the IC and the FETs are OK. You now need to confirm that the resistance of all of the gate drive resistors is correct. If you can read the resistance value on one unburned gate resistor, all should match that value (likely between 22 and 75 ohms). You MUST confirm that each gate resistor is within tolerance (plus/minus 5% of the rated value). If you have a single open resistor, that FET will destroy ALL of the newly replaced FETs when the amplifier is powered up.

If you have ~1/2 B+ at the chip on pins 9 and 10, but did not have it on either set of gate pads, the drivers are likely defective. If the voltage was higher on the pads, it's also likely that the drivers are defective. The drivers are usually set up in an emitter-follower pair (sometimes a diode and PNP transistor on budget amplifiers). It's typical that the PNP transistor is defective. The part number generally begins with 'A' or 'B' if they're using Japanese part numbers. If they're using Motorola type numbers, the PNP transistors are likely MPSA56s. Since the transistors are cheap, it's best that you replace both the NPN and the PNP drivers.

After replacing the drivers (if they were defective) and after confirming that the gate drive resistors are OK (or after replacing all of the defective resistors), the voltage on the gate pads should now read approximately the same as it did on pins 9 and 10 of the IC.

Now you can replace the FETs. ALL FETs MUST come from the same batch (so that they match electronically). If you don't have enough to replace all of them, order a new set of FETs.

You know that the FETs are connected in two opposing banks (one bank to drive each half of the primary winding). The gate resistors in each bank are connected together on one end. The other end of the resistor connects to the FET gate. To be absolutely sure that all FETs are properly connected to the drive circuit, set the meter to resistance and touch one meter lead to the gate of one FET in one bank of FETs. Leave that lead there and touch the other lead to each of the other gates (the gate leg of the FET, not the point where it connects to the board). The resistance in each case should be twice the value of a single resistor. This verifies 100% that the connections AND the resistors are OK. This is VERY important and there is no room for error. If you fine that any reading is significantly different than the others, you have a bad connection or a bad resistor. You MUST find it before applying power. Failure to do so destroys all new FETs.

After the FETs have been replaced, mount the FETs down to the heatsink (insuring that the insulators have no defects between the tab of the transistor and the heatsink). You now need to apply power to the amplifier. Do NOT connect the amplifier directly to the power supply. Virtually any amplifier will power up with a 10 amp fuse. Connect a 10 amp fuse between the amp and the supply and then apply remote power. If the fuse evaporates, there is still a serious problem. Do not go to a larger fuse. That will simply destroy perfectly good FETs. It takes less than ONE second to destroy the FETs if there is a serious problem.

To limit the current, you need to insert a high power resistor (1-2 ohms, 50 watt - shouldn't cost more than ~$5-10US - or the equivalent). Connect a 12 volt lamp across the resistor and connect the resistor/lamp combination in place of the fuse. Now power up the amp (only for a few seconds). If there is a short, the lamp will light. Keep in mind that the resistor will get hot when current is drawn through it. Don't keep the amp powered up too long where the resistor gets dangerously hot. The reason for this setup is to allow you to check for a shorted transformer. Many times the enamel on the transformer's windings will wear through and will allow the windings to short. If this is the case, you can twist the transformer and the current draw will change significantly. If this is the case with your amp, you will need to find the short. I've seen this happen lots of times when the amplifiers were mounted on speaker enclosures.

If the current draw doesn't change when you twist the transformer, it's likely that you have some more problems in the audio section. You will have to go through and check each transistor in the audio section to see if it's shorted. It's time consuming but there is little choice.

Although it's rare, you may want to unsolder the rectifiers (one at a time) and check them for shorts. In the board, they will appear to be shorted between 2 of the pins.

If you find all of the problems and get the amp to power up through the resistor, you must mount it back into the heatsink (remember to check the insulators) before you run audio through it. Outside of the heatsink the transistors will get dangerously hot in seconds (sometimes, even if you are not running audio through it). Mounting it in the heatsink is time consuming but will prevent further damage.

If you get it to play and have it mounted in the sink, you can then play some audio through it. Use a small fuse (a 20 amp fuse will be OK for low to moderate audio levels if the amp is OK). If the amp plays for 10 to 15 minutes and nothing shows signs of overheating, you can be relatively sure that it's properly repaired. Before you go back to the recommended fuse, use 1/2 of the recommended fuse and play it at low to moderate levels for a couple of hours. If it passes that test, it's likely fine.
 
Re: problems with an amp

Perry Babin said:
Hi Jarolda,

You've received a lot of good responses but the following may be helpful (and possibly long and boring) in repairing the amplifier.

If the power supply continues to blow FETs, the problem 'could' be in the power supply but it may be in the audio section. You need to check the audio output transistors. If any output transistor is shorted between pins 2 and 3, the transistor is defective. If the protection circuits in the amplifier aren't working properly (or don't exist), the shorted output transistor would cause the audio section to draw a significant amount of current. This could blow the power supply even if you've replaced all of the defective components. Any defective output transistors must be replaced. If only one output transistor of a bank of parallel transistors is blown, you may remove only the defective transistor to continue with the testing. When you get the amp to a point where it powers up and plays, all parallel transistors (in parallel with the defective one) will have to be replaced.

If the audio output transistors are not shorted, remove all of the FETs from the power supply and connect the amp to the power supply via a relatively small fuse (10 amp should be fine). With the amplifier powered up (including remote) measure the DC voltage on pins 9 and 10 of the power supply IC. They are the last two pins on the top side of the chip on the right side (assuming that the chip is a TL594 or some variant). You should see approximately 1/2 of the B+ input voltage. If you don't, you need to measure the DC voltage on all of the pins and then post them. Look up the datasheet for the chip so you can get the correct number for each pin.

Note: Using a multimeter on 'DC volts' to determine whether the IC is running (producing pulses) is not 100% accurate. Since it's unlikely that you have an oscilloscope, if your meter has a frequency counter, check the frequency on 9 and 10. If you read something near 25khz, you know for sure that the IC is producing pulses.

If the power supply IC seems to be producing output, you then need to check the voltage on the gate pad (where pin 1 of the FET solders into the board). With the black meter lead on ground, the voltage should be approximately the same as you read on the chip. Do this on ALL of the gate pads. If they have ~1/2 voltage on them, then it's likely that the driver transistors between the IC and the FETs are OK. You now need to confirm that the resistance of all of the gate drive resistors is correct. If you can read the resistance value on one unburned gate resistor, all should match that value (likely between 22 and 75 ohms). You MUST confirm that each gate resistor is within tolerance (plus/minus 5% of the rated value). If you have a single open resistor, that FET will destroy ALL of the newly replaced FETs when the amplifier is powered up.

If you have ~1/2 B+ at the chip on pins 9 and 10, but did not have it on either set of gate pads, the drivers are likely defective. If the voltage was higher on the pads, it's also likely that the drivers are defective. The drivers are usually set up in an emitter-follower pair (sometimes a diode and PNP transistor on budget amplifiers). It's typical that the PNP transistor is defective. The part number generally begins with 'A' or 'B' if they're using Japanese part numbers. If they're using Motorola type numbers, the PNP transistors are likely MPSA56s. Since the transistors are cheap, it's best that you replace both the NPN and the PNP drivers.

After replacing the drivers (if they were defective) and after confirming that the gate drive resistors are OK (or after replacing all of the defective resistors), the voltage on the gate pads should now read approximately the same as it did on pins 9 and 10 of the IC.

Now you can replace the FETs. ALL FETs MUST come from the same batch (so that they match electronically). If you don't have enough to replace all of them, order a new set of FETs.

You know that the FETs are connected in two opposing banks (one bank to drive each half of the primary winding). The gate resistors in each bank are connected together on one end. The other end of the resistor connects to the FET gate. To be absolutely sure that all FETs are properly connected to the drive circuit, set the meter to resistance and touch one meter lead to the gate of one FET in one bank of FETs. Leave that lead there and touch the other lead to each of the other gates (the gate leg of the FET, not the point where it connects to the board). The resistance in each case should be twice the value of a single resistor. This verifies 100% that the connections AND the resistors are OK. This is VERY important and there is no room for error. If you fine that any reading is significantly different than the others, you have a bad connection or a bad resistor. You MUST find it before applying power. Failure to do so destroys all new FETs.

After the FETs have been replaced, mount the FETs down to the heatsink (insuring that the insulators have no defects between the tab of the transistor and the heatsink). You now need to apply power to the amplifier. Do NOT connect the amplifier directly to the power supply. Virtually any amplifier will power up with a 10 amp fuse. Connect a 10 amp fuse between the amp and the supply and then apply remote power. If the fuse evaporates, there is still a serious problem. Do not go to a larger fuse. That will simply destroy perfectly good FETs. It takes less than ONE second to destroy the FETs if there is a serious problem.

To limit the current, you need to insert a high power resistor (1-2 ohms, 50 watt - shouldn't cost more than ~$5-10US - or the equivalent). Connect a 12 volt lamp across the resistor and connect the resistor/lamp combination in place of the fuse. Now power up the amp (only for a few seconds). If there is a short, the lamp will light. Keep in mind that the resistor will get hot when current is drawn through it. Don't keep the amp powered up too long where the resistor gets dangerously hot. The reason for this setup is to allow you to check for a shorted transformer. Many times the enamel on the transformer's windings will wear through and will allow the windings to short. If this is the case, you can twist the transformer and the current draw will change significantly. If this is the case with your amp, you will need to find the short. I've seen this happen lots of times when the amplifiers were mounted on speaker enclosures.

If the current draw doesn't change when you twist the transformer, it's likely that you have some more problems in the audio section. You will have to go through and check each transistor in the audio section to see if it's shorted. It's time consuming but there is little choice.

Although it's rare, you may want to unsolder the rectifiers (one at a time) and check them for shorts. In the board, they will appear to be shorted between 2 of the pins.

If you find all of the problems and get the amp to power up through the resistor, you must mount it back into the heatsink (remember to check the insulators) before you run audio through it. Outside of the heatsink the transistors will get dangerously hot in seconds (sometimes, even if you are not running audio through it). Mounting it in the heatsink is time consuming but will prevent further damage.

If you get it to play and have it mounted in the sink, you can then play some audio through it. Use a small fuse (a 20 amp fuse will be OK for low to moderate audio levels if the amp is OK). If the amp plays for 10 to 15 minutes and nothing shows signs of overheating, you can be relatively sure that it's properly repaired. Before you go back to the recommended fuse, use 1/2 of the recommended fuse and play it at low to moderate levels for a couple of hours. If it passes that test, it's likely fine.


Wow!
Thanks for this complete report, I guess I'll have it printed out, I want to make sure I complete every step you describe above.

I haven't done anything to the amp yet, but I have plenty of advices and information now.

Thank you all for the information, tips and advices posted, I'll keep you informed of the progress.

I'm waiting for my digital cam to have some pitures uploaded so you can take a look at it too, it's just a matter of time.

Again, thank you all!! :)