Car Audio Amplifier Repair - JBL GT5-A604

[dinger5]

So I blew my car amplifier (JBL GT5-A604) up trying to drive to low of an impedance. I was driving a 2ohm load for a few hours before it blew up. Minimum apparently was 4ohm for a reason. It was a cheap amp so I don't really care if I can fix it or not, mainly a learning project. I have run out of ideas on what to test to get this thing working again and have now turned to the internet for some help.

Here's a link to the schematics: http://elektrotanya.com/jbl_gt5-a604_sm.pdf/download.html (pdf link is in the middle of the page under the preview)

The power mosfets Q54 and Q53 were completely blown and charred. I replaced all four power mosfets Q44, Q43, Q53, Q54 (Page 17). Plugged it back into car (no power supply yet) and it blew the amplifiers fuses. I have my speakers working with another amp currently so I know it's not a speaker wiring issue.

R44, R43, R53 and R54 all read around 47ohm.

C03, C04, C05 are still reading their correct values.

I think D41 and D42 were still functioning correctly, it does look like the traces are a little charred around these dual diodes though.

I have been hunting for shorts and bad caps but there is nothing obvious to me yet. There is a short from ground to the power rails through somewhere though...

If anyone knows what else I should check besides these power mosfets, help would be greatly appreciated.

[orbanp]

Hi Dinger,

You are lucky that you found the service manual for your amp, with that it is fully repairable, there is nothing exotic in that amp.

Chances are that some of the final transistors are blown and/or the driver transistors, Q134, Q145, Q235, Q245, and Q134, Q144, Q234, Q244.
Do check the same transistors of the other two channels too!

But do go through the whole amp and check it out.

Use a bench power supply (PS) with current limit for the testing, not a battery! If you do not have a PS like that, use a car battery with some light bulb connected in series with it, it will limit the current if you still have some dead short. Use a car bulb, 12V, about 20W - 30W, the brake lights and directional lights have that rating.

Do check the power supply, that inverter with the MOS-FETs. From the power rating of your amp that inveter provides about +/-30V supply to the amps.
Disconnect the power stage from the inverter (VCC1, VEE1) and see if the inverter works properly.
If I am correct, those PSs with the TL494 IC like to have some load on them to start up and operate, so add a light bulb between VCC1 and VEE1, that is a bout 60V, use 120V incandescent light bulb, 40W or so.
If the inverter does not work, do check the MOS-FETS for shorts, also do check the driver transistors, Q23, Q33. Do look up the data sheet of the TL494, there is lots of info in it about its operation.
If the TL494 is damaged, you can harvest one from an old PC power supply.
A scope would not hurt for debugging.

If the inverter works, you can move on to check the power amps. Here is a good article (and web-site) on how to debug power amps: <http://sound.westhost.com/troubleshooting.htm>
The bottom line is that it is good to check all the transistors in it, as the whole circuit is in overall feed back.
The final amp itself is a LIN-type power amp (just like an op-amp), only difference is that it uses a fully symmetrical input stage. The "Leach-amps" used that topology, you can read about that topology in details here:
<http://users.ece.gatech.edu/~mleach/lowtim/>

Good luck, Peter

[orbanp]

Just a quick note.

Do check and replace the final transistors in your amp, the factory devices are woefully undersized!
The 2SC4477 (the NPN side) is a 120V 8A device. That is not enough! (I have not looked at the PNP side, but most probably has the same rating.)

60W output power with 4 Ohm load gives about 5.5A peak current. The speaker is not a resistive load, it has a complex impedance. It introduces phase shift and hence the peak current in practice is a lot higher.

As a rule of thumb, the final transistor should have a voltage rating of about 120% - 130% of the rail-to-rail voltage, while the maximum current rating should be 2 to 3 times higher than the peak current calculated for resistive load. That requires something like a 100V 15A transistor for your amp.

Plugging that data into the Canadian DigiKey search engine, it gives the BD911/912 transistors:

http://www.digikey.ca/product-detail/en/stmicroelectronics/BD911/497-6659-5-ND/1037699
http://www.digikey.ca/product-detail/en/stmicroelectronics/BD912/497-7179-5-ND/1037700

Those are C$1.50 and C$1.80 devices, cheap and available.
Only thing is that the original transistors have TO3P cases while the BD transistors have TO220 cases.
As the TO220 case is smaller, it can be used in the amp without too much trouble.

Peter

[dinger5]

Thank you for the information!
I will try again soon. I will be ordering a power supply and when that comes in I'll be able to debug better. I also have a scope so it should be possible to figure this out.