Got a Rockford Fosgate 400a4. Was told it works but gets hot. Bias pots looked about normal, but sure enough it starts up and draws about 5A at idle. Turned all bias pots fully counterclockwise and idles around 0.80A at startup.
If I attempt to increase the bias pot on any channel, I will get close to the original position and then the current draw will jump about 0.10A and I'll get about 0.003V across a source resistor on that channel. Turning the bias pot counterclockwise won't lower the current or drop the voltage until I get back to fully CCW. This seems to happen on any channel, and it seems sometimes turning the bias back fully CCW won't drop the initial 0.10A current or the 0.003V across the source resistor, until I shut the amp off and restart.
Any ideas?
If I attempt to increase the bias pot on any channel, I will get close to the original position and then the current draw will jump about 0.10A and I'll get about 0.003V across a source resistor on that channel. Turning the bias pot counterclockwise won't lower the current or drop the voltage until I get back to fully CCW. This seems to happen on any channel, and it seems sometimes turning the bias back fully CCW won't drop the initial 0.10A current or the 0.003V across the source resistor, until I shut the amp off and restart.
Any ideas?
This generally happens on channels 3 and 4 but it's been seen before. Check for high-frequency oscillation on the bridging speaker terminals for the channel you're adjusting the bias on.
Does it do it if you reduce the gain on the channels?
Does it do it if you reduce the gain on the channels?
I have the gains at zero. Here is what the right rear channel looks like with the bias all the way CCW, and then when I turn it up until it starts pulling 0.10A above idle. If I adjust the bias from here up or down it'll do a few different things on the scope. This is gains at 0, nothing hooked up to amp besides power & turn-on.
Does it do this with the crossover set to low pass?
How many channels do this?
Does the waveform remain as it is in the second image if the bias is reset to full CCW without cycling power?
How many channels do this?
Does the waveform remain as it is in the second image if the bias is reset to full CCW without cycling power?
Yes, does it with crossover on LP. Had it on AP but switched to LP and it does the same. It gets "worse" when I try to turn the bias back down without power cycling. It's doing it on channels 3 & 4. Channels 1 & 2 don't have the weird output, but still jump ~0.10A when trying to set the bias. Voltage across Ch 1 & 2 source resistors will go to ~0.003V along with the current jump, but will return to normal idle current & voltage when bias is turned back down, whereas Ch 3 & 4 won't return to base idle current & source resistor voltage when bias is turned down.
Ch3 at idle:
Ch3 when bias first increases idle draw 0.10A:
Ch3 with bias reset to 0 CCW:
Ch4 at idle:
Ch4 when bias first increases idle draw 0.10A:
Ch4 with bias reset to 0 CCW:
Ch3 at idle:
Ch3 when bias first increases idle draw 0.10A:
Ch3 with bias reset to 0 CCW:
Ch4 at idle:
Ch4 when bias first increases idle draw 0.10A:
Ch4 with bias reset to 0 CCW:
At this point, no one has taken this beyond knowing what's causing the current draw, as far as I know. I think the changing of the value of a couple of capacitors could resolve this but it would require additional testing with a 20-20kHz sweep to confirm that the amp is still working as it should.
The caps would be Cx13 and Cx15 in the problem channels. The value would have to be incrementally increased or decreased in value (keep the values equal in both locations). I don't know which will help (maybe neither but that's where I would start).
The caps would be Cx13 and Cx15 in the problem channels. The value would have to be incrementally increased or decreased in value (keep the values equal in both locations). I don't know which will help (maybe neither but that's where I would start).
Are you saying it's an inherent flaw in the design/build? I believe I have another same model. I can dig it out and see if it behaves similarly
I think it is a defect. The amp uses the same circuit for all channels so it may be an issue with layout on the board.
Try the other amp but this has happened repeatedly with the 200 and 400a4.
Before you make any mods, solder a short jumper between the RCA grounds for the front and rear channels to see if that makes a difference.
Try the other amp but this has happened repeatedly with the 200 and 400a4.
Before you make any mods, solder a short jumper between the RCA grounds for the front and rear channels to see if that makes a difference.
I found another 400a4, which appeared to be unopened and working normally. It was also biased high and drawing about 3A at idle, so I reset the bias and was able to set it as normal using 0.05A increments (this gives me about 0.001V across source resistors). This gets the idle current down to about 1.1A on this 2nd 400a4. Aside from Ch3, I get about the same response if I probe either + or - on each channel. These are the bridgeable terminals with the scope grounded to B- of this second 400a4 for comparison.
Ch1
Ch2
Ch3
Ch4
Ch1
Ch2
Ch3
Ch4
Is the board the same number?
Are you saying that ch3 on the second amp is doing the same as the first amp?
Ground the scope probe to the non-bridging terminal of the channel you're probing.
Are you saying that ch3 on the second amp is doing the same as the first amp?
Ground the scope probe to the non-bridging terminal of the channel you're probing.
I thought the boards were the same, PC-1983, but after cleaning some thermal paste off this one is a "C" and the first is an "E".
Ch3 is similar to the first amp, but with this second amp I can set all four channel biases normally without issue unlike the first amp. Each will let me gradually adjust the bias to 0.05A without jumping, and will drop back down if I turn CCW without having to power cycle.
Here are the 4 channels with scope grounded to non-bridging terminals of each channel:
Ch1
Ch2
Ch3
Ch 4
Changed horizontal...
Ch1
Ch2
Ch3
Ch4
Ch3 is similar to the first amp, but with this second amp I can set all four channel biases normally without issue unlike the first amp. Each will let me gradually adjust the bias to 0.05A without jumping, and will drop back down if I turn CCW without having to power cycle.
Here are the 4 channels with scope grounded to non-bridging terminals of each channel:
Ch1
Ch2
Ch3
Ch 4
Changed horizontal...
Ch1
Ch2
Ch3
Ch4
I don't have any other suggestions other than trying different values for the two capacitors for the two bad channels.
Do you have any thoughts as to why the first amp won't bias normally, or you think that's tied to the oscillation?
Yes. The amps have a limited design bandwidth. That's normal for audio amplifiers. The high-frequency oscillation is outside of what it was designed for and the circuit can't react quickly enough to prevent the opposing output FETs from conducting at the same time.
For some of the Rockford amps (some of the ones with the 'flat' bass and treble pots) will oscillate (gains and treble set to high settings) and cause enough cross-conduction of the outputs (drawing very high current) to cause the outputs to fail.
For some of the Rockford amps (some of the ones with the 'flat' bass and treble pots) will oscillate (gains and treble set to high settings) and cause enough cross-conduction of the outputs (drawing very high current) to cause the outputs to fail.
I'll have to double check, but I don't think Ch 1 & 2 cause any oscillation on Ch3, so I'm wondering why they also won't bias normally.
What makes Ch 3 & 4 susceptible to the oscillation when the individual channel schematics seem to be about identical between all 4 channels?
What makes Ch 3 & 4 susceptible to the oscillation when the individual channel schematics seem to be about identical between all 4 channels?
It could be that they used a ground for some part of the circuit that wasn't clean/solid.
Other than replacing the caps I suggested, another solution may to be use a different transistor in one location, or another. I built some amps that worked well but were a bit noisier (not audible) than a larger version. The solution was to go from an MPSA06/56 to an MPSA42/92. It's sometimes difficult to predict whats going to happen with a given layout of the board.
Other than replacing the caps I suggested, another solution may to be use a different transistor in one location, or another. I built some amps that worked well but were a bit noisier (not audible) than a larger version. The solution was to go from an MPSA06/56 to an MPSA42/92. It's sometimes difficult to predict whats going to happen with a given layout of the board.
Still trying to figure this one out. With all bias pots fully CCW, when I check voltage on the IRF540 gates initially and then after warming up ~10 minutes, I get:
Ch1 -30.23 --> -30.20
Ch2 -28.65 --> -28.77
Ch3 -28.62 --> -29.50
Ch4 -28.58 --> -28.72
Upon initial startup, Ch3 reads around -28.62 and I can watch it go to about -29.60 within 30 seconds, settling around -29.50 after a while. The other channels don't creep like this. Also, checking voltage drop across Rx39 after warm up I get:
R139 0.205
R239 1.626
R339 0.875
R439 1.682
I'm in over my head, but thought this was odd enough to mention. Any ideas?
Ch1 -30.23 --> -30.20
Ch2 -28.65 --> -28.77
Ch3 -28.62 --> -29.50
Ch4 -28.58 --> -28.72
Upon initial startup, Ch3 reads around -28.62 and I can watch it go to about -29.60 within 30 seconds, settling around -29.50 after a while. The other channels don't creep like this. Also, checking voltage drop across Rx39 after warm up I get:
R139 0.205
R239 1.626
R339 0.875
R439 1.682
I'm in over my head, but thought this was odd enough to mention. Any ideas?