I just finished rebuilding this Rockford Fosgate amplifier and I have a thread going where I asked about proper bias current. I didn’t get much as far as hits on that thread, Nelson said 30 mA per device is “a healthy start”, so I am going to just set it there. There are 16 MOSFETs per channel so that would be .24 amps per rail, but sadly I can’t measure the rails unless I remove the large blue buss bar which I guess is and option.
So what I did is, I measured across one of the one ohm source resistors and adjusted bias until I saw 30 mV across that resistor which would equal 30 mA. Both channels perfectly even. Well one channel got quite a bit warmer. I checked all of the other source resistors and found that the voltages were all over the place, that’s when I found that these are not like emitter resistors on BJTs. On the cooler channel, the resistor I was measuring on was about the average most of them being in the 30 to 40 mil range while a couple were as high as 50 and some as low as 20 mV. While on the channel that ran warmer, the resistor that measured 30 was actually on of the lowest measuring, most of the resistors on that side were in the 50 to 60 mV range with a few at about 80 mV.
I’ve been doing some reading and I found others. They have had this situation and this shows the importance of matching your MOSFETs. So these people went back to their PB to get better matched sets. Well this is from the factory this way and I’m not gonna be able to better match 32 MOSFETs. I currently don’t have any spares of these models so I would have to buy a whole bunch.
So how would you recommend to best set bias on this? Should I just remove the buss bar from one rail? I could just make sure that the current draw on the positive rail is the same for both channels? I’ve read that the source resistor is generally the way it is done, but how exactly when the measurements are so different from each other. Do I just go for an average? Write down all the measurements I get for each 16 for each channel and average them out?
I really appreciate any help I can get with this. I don’t mess with mosfet too often, be a good learning experience.
Thanks,
Dan
So what I did is, I measured across one of the one ohm source resistors and adjusted bias until I saw 30 mV across that resistor which would equal 30 mA. Both channels perfectly even. Well one channel got quite a bit warmer. I checked all of the other source resistors and found that the voltages were all over the place, that’s when I found that these are not like emitter resistors on BJTs. On the cooler channel, the resistor I was measuring on was about the average most of them being in the 30 to 40 mil range while a couple were as high as 50 and some as low as 20 mV. While on the channel that ran warmer, the resistor that measured 30 was actually on of the lowest measuring, most of the resistors on that side were in the 50 to 60 mV range with a few at about 80 mV.
I’ve been doing some reading and I found others. They have had this situation and this shows the importance of matching your MOSFETs. So these people went back to their PB to get better matched sets. Well this is from the factory this way and I’m not gonna be able to better match 32 MOSFETs. I currently don’t have any spares of these models so I would have to buy a whole bunch.
So how would you recommend to best set bias on this? Should I just remove the buss bar from one rail? I could just make sure that the current draw on the positive rail is the same for both channels? I’ve read that the source resistor is generally the way it is done, but how exactly when the measurements are so different from each other. Do I just go for an average? Write down all the measurements I get for each 16 for each channel and average them out?
I really appreciate any help I can get with this. I don’t mess with mosfet too often, be a good learning experience.
Thanks,
Dan
I went ahead and just pulled the blue buss bar for one of the rails and adjusted for current draw. After a half an hour of warm-up both channels are right at 240 mA which would be 30 per device. Both channels are exactly the same temperature. Now, he sings her just a bit warm sitting at idle. I would still like to learn the proper method of doing it by measuring the voltage drop across the source resistors and how to balance them out as it seems each device pulls a different amount of current. At least this way I know that the channels are balanced.
Dan
Dan