Hi everyone.
I have been doing repairs on these types of amplifiers for a few years, and many times it has happened that after completing the repair, the amplifier whistles a variable frequency whistle (based on battery voltage and how hard it is working). ) that you can hear very well if I use a normal speaker, but you can listen very little if I use subwoofers inside the trunk.
This seems normal to me, but I would like your opinion.
The strange thing is that this whistle is radiated inside the stereo system to which this amplifier is connected and is picked up by the other amplifiers of the audio chain and therefore can be heard in the front speakers of the vehicle.
I thought this happens because the shielding (GND) of the amplifier's RCA cables always coincides with the secondary ground of the transformers, which, not having a switching frequency that matches the switching frequency of the class D stage ( usually the class D stage should oscillate at 4 times the speed of the power supply, but this is never the case) due to this "noise".
Has anyone solved this problem? At least keep the whistle from spreading through the rest of the sound system.
P.s I noticed that the whistle is more consistent the bigger the amplifier in question.
I have been doing repairs on these types of amplifiers for a few years, and many times it has happened that after completing the repair, the amplifier whistles a variable frequency whistle (based on battery voltage and how hard it is working). ) that you can hear very well if I use a normal speaker, but you can listen very little if I use subwoofers inside the trunk.
This seems normal to me, but I would like your opinion.
The strange thing is that this whistle is radiated inside the stereo system to which this amplifier is connected and is picked up by the other amplifiers of the audio chain and therefore can be heard in the front speakers of the vehicle.
I thought this happens because the shielding (GND) of the amplifier's RCA cables always coincides with the secondary ground of the transformers, which, not having a switching frequency that matches the switching frequency of the class D stage ( usually the class D stage should oscillate at 4 times the speed of the power supply, but this is never the case) due to this "noise".
Has anyone solved this problem? At least keep the whistle from spreading through the rest of the sound system.
P.s I noticed that the whistle is more consistent the bigger the amplifier in question.
http://www.bcae1.com/temp/cooleditmixedoscillator01.wav
If it sounds like this ^, it's caused by unsynchronized oscilators (output and PS). The noise can't be fixed (practically) as far as the output of the amplifier is concerned.
The noise transmitted into the system seems to be due to little or no resistance between the primary and secondary grounds in the amp. Others have had this problem. I don't know if they found a solution.
If it sounds like this ^, it's caused by unsynchronized oscilators (output and PS). The noise can't be fixed (practically) as far as the output of the amplifier is concerned.
The noise transmitted into the system seems to be due to little or no resistance between the primary and secondary grounds in the amp. Others have had this problem. I don't know if they found a solution.
Yes Perry, the trouble is just that!
But it seems very strange to me that the problem of the whistle that radiates in the rest of the system is due to low resistance between primary and secondary soil, because if I disconnect the RCAs from the amplifier, the whistle disappears from the system.
As for the resistor (usually 1k) between secondary and primary ground, it is strange, because that resistor, when intact, remains 1kohm until you insert the RCAs, which will connect the secondary ground to the primary ground through the radio (the value resistance of an RCA cable is a few ohms and the shielding of the RCA outputs of a radio is essentially connected with the primary ground, i.e. the ground of the whole car).
The amps for the other person with this problem had a lower value resistor in-between the grounds.
Some people have used ground loop isolators to get around this.
Some people have used ground loop isolators to get around this.
Connecting couple of these amps in a demo or spl car make the whistle almost unavoidable plus when You add some alternator noise and/or a ground loop issue...it's a nightmare to deal with.
What I've noticed is the more the amps in a system, the more the noise...
Ground loop isolators - affect the sound in a very very negative way
What I've noticed is the more the amps in a system, the more the noise...
Ground loop isolators - affect the sound in a very very negative way
Quality transformer based ones do not in my experience, especially if used to isolate the big class D amp from the rest of the system rather than using them on the full range amps.
Some of the isolators are directional and if connected in reverse (because the RCA connectors aren't right for your system, without adapters), they can reduce the bass.
It could be possible to re-wire the input circuit to convert it to a quasi-balanced input to get rid of the noise but, for my business, I didn't do any modifications unless they were reliability related. If the owner bought a poorly designed amp, they received that amp precisely as it left the factory when I repaired it. If they weren't happy with it, they should have returned it for something else.
It could be possible to re-wire the input circuit to convert it to a quasi-balanced input to get rid of the noise but, for my business, I didn't do any modifications unless they were reliability related. If the owner bought a poorly designed amp, they received that amp precisely as it left the factory when I repaired it. If they weren't happy with it, they should have returned it for something else.
Guys, honestly I would like to try to isolate the amplifier from the rest of the system using a pair of OP AMPS (for example tl072), in this way, it seems that the shielding of the rca coming from the radio does not directly touch the shielding of the amplifier , this way you should stop the noise.
But if I'm wrong, correct me.
thanks.
But if I'm wrong, correct me.
thanks.
Attachments
It should help prevent the noise from traveling back through the system but it won't stop the amp from having the noise from the un-synchronized oscillators.
It's a lot more work than a GLI since you have to supply power to the op-amps.
It's a lot more work than a GLI since you have to supply power to the op-amps.
Yes Perry, I agree, GLI is ready to be interposed between the RCA cables, this circuit instead must be powered.
It shouldn't be a problem, because usually these amplifiers have the preamp section on a separate board that literally detaches from the rest of the amplifier, it can slide from the chassis even without removing the amplifier from the car, so the circuit can be integrated (interposed ) working only on the pre card.
I can "hook" the power supply (+/- 15v) on pins 8 and 4 of one of the many OP AMPS already existing on the board and earn the secondary GND simply by hooking to the RCA shield of the amplifier itself.
Also (correct me if I'm wrong) a similar solution does not degrade or modify or alter the signal (especially the ultra-low frequencies) as could happen with a GLI.
I know very well that any solution adopted will not prevent the amplifier from whistling, but at least these whistles will not fit into the rest of the system, right?
Last edited:
You may have to experiment with the ground.
Sometimes the best solution is the simplest solution.
Did you attempt to test one or more GLIs to determine precisely how much they decrease the low frequency response? Remembering that some have to be oriented properly in the signal path, regarding input and output?
I'm not trying to discourage you from building the circuit but is it worth it to lay out a board for a specific case, to get boards printed and assemble them, pay for the case for it and then install them?
I've been through the process of having to build various circuits and, for me, it's far too much trouble, especially when you can simply tell the owner to go buy a $15 part. Can you sell the device for that price?
If you want to modify the amp, I'd look at simply making a slight mod to the circuit that's there. It may mean nothing more than breaking the ground from the RCA shields and inserting a 1k ohm resistor, then connecting the audio input ground to the actual RCA shields.
Sometimes the best solution is the simplest solution.
Did you attempt to test one or more GLIs to determine precisely how much they decrease the low frequency response? Remembering that some have to be oriented properly in the signal path, regarding input and output?
I'm not trying to discourage you from building the circuit but is it worth it to lay out a board for a specific case, to get boards printed and assemble them, pay for the case for it and then install them?
I've been through the process of having to build various circuits and, for me, it's far too much trouble, especially when you can simply tell the owner to go buy a $15 part. Can you sell the device for that price?
If you want to modify the amp, I'd look at simply making a slight mod to the circuit that's there. It may mean nothing more than breaking the ground from the RCA shields and inserting a 1k ohm resistor, then connecting the audio input ground to the actual RCA shields.
No perry, unfortunately I did not test with any GLI, but it was enough for me to get informed online and listen to the opinion of some people who have used it.
According to them, the functionality is good, but for trained ears it is evident that part of the musical information in ultra-low frequencies below 50hz are attenuated, this is unpleasant especially because if I use a monster amplifier like that it is because I want a monstrous bass, therefore, if this device kills me down, it makes no sense.
If the device I plan to make works well, I am not going to sell it, but if someone asked me (from that moment) a solution to this problem, I would know how to do it, integrating this circuit inside the amplifier, not with an external box.
I have to be honest, I had already thought about separating the shielding of the rca with a 1k resistor, but then I thought that there will never be a real separation through a simple resistor.
What do you think about it?
I have a few transformer based isolators, in can test them for level and distortion on my scope, it isn’t the best but has FFT that can resolve most gross forms of distortion.
I have to confess that in my own listening tests, including some high end subs and amplifiers, that it doesn’t make much difference.
I shall see if I can provide the data to back it up.
I have to confess that in my own listening tests, including some high end subs and amplifiers, that it doesn’t make much difference.
I shall see if I can provide the data to back it up.
3rd opinion here, I would encourage you to at least try a GLI. I used quite a few over the years, and was never disappointed. None were ever removed for audibly degrading performance.
I did a couple of op amp circuits years ago for adapting differential inputs/outputs within car systems and was disappointed that all the effort didn't work better than the simple pair of transformers in a ground loop isolator.
I did a couple of op amp circuits years ago for adapting differential inputs/outputs within car systems and was disappointed that all the effort didn't work better than the simple pair of transformers in a ground loop isolator.
#12:
If you want the worst information possible, go with what everyone is telling you online. You have to test for yourself or take the results from someone who is reliable.
I don't know for certain what it's going to take to get rid of the noise. The solution may be simply moving a couple of connections or it may take more. If you want to do something more complex for the knowledge, I'm not against that. Again, for me, the best solution is the simplest solution.
#13:
If you test isolators, you need to test the relatively large cylindrical ones with RCAs that come out of each end.
If you want the worst information possible, go with what everyone is telling you online. You have to test for yourself or take the results from someone who is reliable.
I don't know for certain what it's going to take to get rid of the noise. The solution may be simply moving a couple of connections or it may take more. If you want to do something more complex for the knowledge, I'm not against that. Again, for me, the best solution is the simplest solution.
#13:
If you test isolators, you need to test the relatively large cylindrical ones with RCAs that come out of each end.