Hi all.
I use to repair lots of car audio amps, probably 5-10 a month but this one I really cracked my head..
The failure is at LEFT channel and the problem is that the negative side of the output signal is clipping abnormal.
I spent a few hours of logical thinking and some measurements, without finding anything abnormal.
The amp is not loaded at all, I do not connect any of my dummy loads before the amp operates correctly in open load.
Anyways, I've changed ALL small signal SMD transistors, tested with a pair of new output FET's, made resistance measurements to all SMD resistors, everything is correct.
Also, DC levels during idle is the same as the OK channel.
The only difference I can find is, the voltage amplifier stage Q109 and Q111, doesn't give a clean negative swing. It stays a volt or so negative. This compared to the OK channel does drive -13V negative. This is when driving the amp to clipping.
Anyone have any ideas?
I use to repair lots of car audio amps, probably 5-10 a month but this one I really cracked my head..
The failure is at LEFT channel and the problem is that the negative side of the output signal is clipping abnormal.
I spent a few hours of logical thinking and some measurements, without finding anything abnormal.
The amp is not loaded at all, I do not connect any of my dummy loads before the amp operates correctly in open load.
Anyways, I've changed ALL small signal SMD transistors, tested with a pair of new output FET's, made resistance measurements to all SMD resistors, everything is correct.
Also, DC levels during idle is the same as the OK channel.
The only difference I can find is, the voltage amplifier stage Q109 and Q111, doesn't give a clean negative swing. It stays a volt or so negative. This compared to the OK channel does drive -13V negative. This is when driving the amp to clipping.
Anyone have any ideas?
I'm waiting on a schematic from Rockford.
The problem is possibly an open resistor in the protection circuit. Sometimes it's easy to miss them because they're in parallel with the low value source resistors.
Are all of the source resistors within tolerance?
Did any of the source resistors fail initially?
The problem is possibly an open resistor in the protection circuit. Sometimes it's easy to miss them because they're in parallel with the low value source resistors.
Are all of the source resistors within tolerance?
Did any of the source resistors fail initially?
Double-check R161, R162 and R163. You may have to desolder one leg of R163 to get an accurate reading on them.
You can disable the protection circuit by lifting D102 and D104. If the amp works properly with those out of the circuit, you know that the audio circuit is OK and the problem is with the protection circuit.
You can disable the protection circuit by lifting D102 and D104. If the amp works properly with those out of the circuit, you know that the audio circuit is OK and the problem is with the protection circuit.
I found the schematics and uploaded it:
http://www.etanoldax.se/RF201s.pdf
The source resistors is OK. The amp behaves like this even without load, with low output current.
I tried to disable the over current protection circuit, but it didn't do any difference.
Please take a look in the schematics, I'm out of ideas....
http://www.etanoldax.se/RF201s.pdf
The source resistors is OK. The amp behaves like this even without load, with low output current.
I tried to disable the over current protection circuit, but it didn't do any difference.
Please take a look in the schematics, I'm out of ideas....
R163 measures like all the others. R162 is not mounted, and R161 is a zero ohm "resistor" and measures zero aswell.
This is the same on the OK channel so probably Rockford made some changes after the schematics were printed.
Like I said in my first post, collector Q111 doesn't go much negative when output calls for maximum swing (when driving into clip). Q111 should be saturated during this condition, and does on the OK channel. It goes only a volt or two negative while the OK channel goes -13V.
This leads to a maximum gate-source (Q115) voltage of 5V, instead of 17V on the OK channel.
I already tried lifting D102 and D104. Helps Nothing!
This is the same on the OK channel so probably Rockford made some changes after the schematics were printed.
Like I said in my first post, collector Q111 doesn't go much negative when output calls for maximum swing (when driving into clip). Q111 should be saturated during this condition, and does on the OK channel. It goes only a volt or two negative while the OK channel goes -13V.
This leads to a maximum gate-source (Q115) voltage of 5V, instead of 17V on the OK channel.
I already tried lifting D102 and D104. Helps Nothing!
Well, that's about what I have on a similar amp (335mv across the 47 ohm resistors).
Does the emitter of 111 follow the base (minus ~0.6v)? Check it loaded and unloaded. Compare to the other channel.
Does the amplitude of the base drive for 111 compare to the other channel (loaded and unloaded)?
Does the emitter of 111 follow the base (minus ~0.6v)? Check it loaded and unloaded. Compare to the other channel.
Does the amplitude of the base drive for 111 compare to the other channel (loaded and unloaded)?
There is some hokus pokus around Q111. As I tried to explain in a previous post, the collector voltage at Q111 doesn't drive the negative side of output stage very much. I compare both channels when applying a 1.5kHz sinewave, driving both channels a bit above clip so I can see the feedback loop saturating in either direction. It's obvious a big difference to the negative going saturation of Q111, it will just go one or two volts negative compared to positive about +13V. The OK channel goes symmetrical to +13V and -13V during output clipping (feedback loop saturated)
Also, as long as the output is held from clipping there are very small voltage swings at collector Q111. Only when clipping there are high swings (just like a square wave). The failing channel does have a higher voltage swing at collector Q111 at all times, make me believe the loopgain is lower.
It looks like Q111 doesn't get a proper drive level, preventing it from saturation.
Also, as long as the output is held from clipping there are very small voltage swings at collector Q111. Only when clipping there are high swings (just like a square wave). The failing channel does have a higher voltage swing at collector Q111 at all times, make me believe the loopgain is lower.
It looks like Q111 doesn't get a proper drive level, preventing it from saturation.
Connect a dummy load to the amp and drive it to moderate clipping. Compare the voltage across the resistors in series with the transistors.
For example, measure the voltage across Rx24 in both channels. If there is more voltage across the one in the defective channel, it means that the amp is working harder to compensate for the error. If there is less voltage drop, determine why there is insufficient current passing through the resistor.
If you see more voltage across R146 than 246, you know that the circuit is trying to pull the voltage negative but something is loading the collector of Q111 (possibly low gain for Q113?).
Work your way through the circuit.
For example, measure the voltage across Rx24 in both channels. If there is more voltage across the one in the defective channel, it means that the amp is working harder to compensate for the error. If there is less voltage drop, determine why there is insufficient current passing through the resistor.
If you see more voltage across R146 than 246, you know that the circuit is trying to pull the voltage negative but something is loading the collector of Q111 (possibly low gain for Q113?).
Work your way through the circuit.
Hello Perry, sorry for the delay in getting back.
I'm still stuck with this amp...
I measured the voltage across R124, it seems like when going into clipping, the voltage drops. So, obviously the differential input stage will not drive the voltage amp stage hard enough.
The OK channel, also drops the voltage across R246, during clip, but not all the way like the failing ch.
I took some pictures of the oscilloscope during some measurements.
The above picture shows the signal at output terminal (high amplitude) and collector Q111. It can be seen that during negative clip, something strange happens to Q111. Positive swing seems OK though..
This picture shows how nice looking waveform it's at the OK channel, collector Q211.
During hard clipping, it's obvious the voltage amplifier stage gives too less drive for the output stage FET's.
I'm completely out of ideas ...
Regards
Jackson
I'm still stuck with this amp...
I measured the voltage across R124, it seems like when going into clipping, the voltage drops. So, obviously the differential input stage will not drive the voltage amp stage hard enough.
The OK channel, also drops the voltage across R246, during clip, but not all the way like the failing ch.
I took some pictures of the oscilloscope during some measurements.
The above picture shows the signal at output terminal (high amplitude) and collector Q111. It can be seen that during negative clip, something strange happens to Q111. Positive swing seems OK though..
This picture shows how nice looking waveform it's at the OK channel, collector Q211.
During hard clipping, it's obvious the voltage amplifier stage gives too less drive for the output stage FET's.
I'm completely out of ideas ...
Regards
Jackson
It may be easier to troubleshoot this from a DC point of view.
I tried this on the amp I have here. I used a 50k ohm potentiometer. One outer leg of the pot was connected to +15v. The other outer leg was connected to -15v. The wiper was connected to the input of the amp.
So that you can measure the voltage across the various resistors with a multimeter, drive the input (Q102) with the wiper of the pot. The voltage on the output should follow the wiper voltage precisely. If it doesn't, you can use your multimeter to measure the voltage on the various resistors to see where the problem is.
You'll likely have to remove C123 and D101 to prevent them from interfering with the drive voltage.
If you want to drive it to the rails (beyond the ±15v from the regs), solder a jumper in place of C123. Without the jumper, the gain of the amp is 1. With the jumper, the gain is 6.
Before you connect the pot, you may want to try jumping C123 and removing D101 to see if it makes a difference.
I tried this on the amp I have here. I used a 50k ohm potentiometer. One outer leg of the pot was connected to +15v. The other outer leg was connected to -15v. The wiper was connected to the input of the amp.
So that you can measure the voltage across the various resistors with a multimeter, drive the input (Q102) with the wiper of the pot. The voltage on the output should follow the wiper voltage precisely. If it doesn't, you can use your multimeter to measure the voltage on the various resistors to see where the problem is.
You'll likely have to remove C123 and D101 to prevent them from interfering with the drive voltage.
If you want to drive it to the rails (beyond the ±15v from the regs), solder a jumper in place of C123. Without the jumper, the gain of the amp is 1. With the jumper, the gain is 6.
Before you connect the pot, you may want to try jumping C123 and removing D101 to see if it makes a difference.
Hi Perry!
I tried the DC troubleshooting. I connected a pot according to your instructions.
There is some strange behaviour on the negative side..
By the way, it seems the gain is about 2 without shorting C123 since the output stage gives a gain of 2. Anyway, when turning the pot, applying negative DC at Q102 base, the output starts to follow correctly. But it's restricted to 4 volts negative...
Positive is all ok, right up to the rail.
If turning the pot fast from zero till full negative, the output follows, and for 1/10 second the output follows full negative but returns to -4 volts again.
When applying negative input, it seems most of the current disappears from R124.
I can't understand. The input stage should really drive Q104/Q105 (current mirror) hard enough to drive Q111. Also, I measure voltage drop across R121, which goes next to zero aswell.
The current source seems to be somewhat constant, measuring the drop across R122.
It's waaaaay crazy!!
I tried the DC troubleshooting. I connected a pot according to your instructions.
There is some strange behaviour on the negative side..
By the way, it seems the gain is about 2 without shorting C123 since the output stage gives a gain of 2. Anyway, when turning the pot, applying negative DC at Q102 base, the output starts to follow correctly. But it's restricted to 4 volts negative...
Positive is all ok, right up to the rail.
If turning the pot fast from zero till full negative, the output follows, and for 1/10 second the output follows full negative but returns to -4 volts again.
When applying negative input, it seems most of the current disappears from R124.
I can't understand. The input stage should really drive Q104/Q105 (current mirror) hard enough to drive Q111. Also, I measure voltage drop across R121, which goes next to zero aswell.
The current source seems to be somewhat constant, measuring the drop across R122.
It's waaaaay crazy!!
I think you'll have to jump C123 to get linear response both positive and negative. For one polarity, the cap blocks DC and gives no DC gain. In the other direction, it will leak and cause you to have more gain. With it out of the circuit, you should see a gain of 1 both positive and negative.
Either jump or remove C123 to see if things change.
Either jump or remove C123 to see if things change.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.