Hi guys I am hoping to finally resolve ground loop noise on my DIY 8-ch. LM3886-based car amplifier. After much debugging and modifications, it's a great sounding amp with blue LED illumination.
However even after isolating all signal grounds from power ground(s) & vehicle, I somehow manage to pick up ground loop noise when an RCA connector from the next pair of outputs from my source (Alpine PXA-H701 DSP/EQ) is used. Until then, (one pair only) no noise.
Looks like I'll have to finally compromise and add a balanced input circuit as Perry Babin & others recommend. Ground loop isolators aren't working as they cause my SMPS fuses to pop when used (amp oscillating?). The amps are PA100 design examples from National Semiconductor.
So my question:
Does anyone have a good balanced input circuit example they know will work for this?
I have been searching quite a bit without so much luck. Also tried Rod Elliot's Project 51, figure 2, balanced input but still noisy.
I guess basically I need to confirm the right approach, and right understanding of how the balanced input circuit can work for noise cancellation even if my input is not balanced.
Thanks as I would really appreciate the help. You can imagine how I feel after working on the amp for months, and thought it was resolved, only to find out when used fully....NOISE!
However even after isolating all signal grounds from power ground(s) & vehicle, I somehow manage to pick up ground loop noise when an RCA connector from the next pair of outputs from my source (Alpine PXA-H701 DSP/EQ) is used. Until then, (one pair only) no noise.
Looks like I'll have to finally compromise and add a balanced input circuit as Perry Babin & others recommend. Ground loop isolators aren't working as they cause my SMPS fuses to pop when used (amp oscillating?). The amps are PA100 design examples from National Semiconductor.
So my question:
Does anyone have a good balanced input circuit example they know will work for this?
I have been searching quite a bit without so much luck. Also tried Rod Elliot's Project 51, figure 2, balanced input but still noisy.
I guess basically I need to confirm the right approach, and right understanding of how the balanced input circuit can work for noise cancellation even if my input is not balanced.
Thanks as I would really appreciate the help. You can imagine how I feel after working on the amp for months, and thought it was resolved, only to find out when used fully....NOISE!
Connecting the secondary center tap to chassis ground may stop the oscillation. The secondary is only left floating in amps that use the isolation of the secondary as a means of preventing ground loops.
Do you have buffers at the input of the amp or do you have the input feeding the amplifier ICs directly?
Do you have buffers at the input of the amp or do you have the input feeding the amplifier ICs directly?
Afternoon Perry,
wouldn't he benefit from the old Rockford Fosgate TOPAZ input design that used on most all their old stuff ? RF had a input design that almost totally isolate the input buffer and crossover section of the amp from the main amp circuits.
Seems like this might help the fella out, and just thinking out loud..
wouldn't he benefit from the old Rockford Fosgate TOPAZ input design that used on most all their old stuff ? RF had a input design that almost totally isolate the input buffer and crossover section of the amp from the main amp circuits.
Seems like this might help the fella out, and just thinking out loud..
Hi Perry & Just1moreamp:
Yes I am driving the ICs directly; you may remember my original thread (direct path to amp modules).
It sounds like if I use ground loop isolators, and connect the grounds as described, I'll introduce a ground loop on all others and thus will need isolators on all 8 RCA inputs. Not very efficient vs. resolving using solid state parts.
The current set-up works quite well until I plug in the next pair additional RCA inputs (The source, the Alpine PXA-H701 is an EQ/Crossover/5.1 surround sound decoder all in one-no other units used). This is true for each of three SMPSs in my amplifier. No noise until next pair is plugged in.
I never had this issue with my older commercial amplifiers.
If anyone can recommend an amp schematic source, or the input circuit schematic, that would be appreciated. Not familiar with TOPAZ, but would love to learn more.
Thanks!
PS: That's my amp in the picture, sounds really good and glows bright blue when powered up. It's been my first "dream project."
Yes I am driving the ICs directly; you may remember my original thread (direct path to amp modules).
It sounds like if I use ground loop isolators, and connect the grounds as described, I'll introduce a ground loop on all others and thus will need isolators on all 8 RCA inputs. Not very efficient vs. resolving using solid state parts.
The current set-up works quite well until I plug in the next pair additional RCA inputs (The source, the Alpine PXA-H701 is an EQ/Crossover/5.1 surround sound decoder all in one-no other units used). This is true for each of three SMPSs in my amplifier. No noise until next pair is plugged in.
I never had this issue with my older commercial amplifiers.
If anyone can recommend an amp schematic source, or the input circuit schematic, that would be appreciated. Not familiar with TOPAZ, but would love to learn more.
Thanks!
PS: That's my amp in the picture, sounds really good and glows bright blue when powered up. It's been my first "dream project."
The rockford circuit uses the shields as a reference for the op-amp power supply which I would prefer to avoid.
The circuit below works. An RC network would probably be useful at the input of both buffers to shunt high frequency noise to ground. It's quiet when using good op-amps. You don't have to use anything expensive. NE5532s will work well. 1% components should be used between the buffers and the next stage of the circuit.
The circuit below works. An RC network would probably be useful at the input of both buffers to shunt high frequency noise to ground. It's quiet when using good op-amps. You don't have to use anything expensive. NE5532s will work well. 1% components should be used between the buffers and the next stage of the circuit.
Attachments
Thank you very much, sir.
Would you mind noting the following, please?:
1. "which I would prefer to avoid" - Why do you say that? What is preferrable?
2. I assume the ground connection shown will go the op-amp power supply source, which is in my case the output of my SMPS-correct? (I use voltage regulators of course)
Thanks!!
Would you mind noting the following, please?:
1. "which I would prefer to avoid" - Why do you say that? What is preferrable?
2. I assume the ground connection shown will go the op-amp power supply source, which is in my case the output of my SMPS-correct? (I use voltage regulators of course)
Thanks!!
Why not just connect the shield direct to the inverting leg of the diff amp, and the signal to the non-inverting leg? i.e. ditch the two front end buffers.
I did this when I had a bad alt whine many years ago and it worked perfect.
Or does your way have better CMRR or something Perry?
I did this when I had a bad alt whine many years ago and it worked perfect.
Or does your way have better CMRR or something Perry?
richie00boy:
I've done this but for some reason there was more noise (noise floor slightly higher) than when I used the buffers.
MartyM:
I think there may be another problem. Having all grounds (except the RCA shield ground) connected to chassis ground has worked for me. I've used the circuit with speaker level and RCA level inputs and it's always been quiet.
Can you post a schematic of the circuit you're using for the audio IC?
I've done this but for some reason there was more noise (noise floor slightly higher) than when I used the buffers.
MartyM:
I think there may be another problem. Having all grounds (except the RCA shield ground) connected to chassis ground has worked for me. I've used the circuit with speaker level and RCA level inputs and it's always been quiet.
Can you post a schematic of the circuit you're using for the audio IC?
Balanced inputs are not a solution to solving ground loop problems.
I know of many situations where commercial amps are used, that have balanced inputs, and they still have ground loop hum.
Balanced lines are simply designed to eliminate (reduce) common mode noise that is picked up by (usually long) cables that feed an amplifier or mixing device.
If you are experiencing hum or earth loop issues, then it's time to look at the design of your power supply, input stages, grounding, isolation and decoupling.
I know of many situations where commercial amps are used, that have balanced inputs, and they still have ground loop hum.
Balanced lines are simply designed to eliminate (reduce) common mode noise that is picked up by (usually long) cables that feed an amplifier or mixing device.
If you are experiencing hum or earth loop issues, then it's time to look at the design of your power supply, input stages, grounding, isolation and decoupling.
In mobile amplifiers, you have to provide some way to ensure that the reference for the audio is the shield of the head unit (or whatever signal source is driving the inputs of the amplifier).
The most common method (used for virtually all budget amplifiers) is to float the secondary of the power supply. This works well as long as all inputs are being driven from the same signal source.
The input circuit that Rockford uses floats the ground for the 15v regulators. This works well but can produce noise unless all signals are from the same source.
Other amps use balanced inputs (or some form of active input circuit). These work best when the inputs of the amp come from different sources (i.e. sub output from head unit directly to the amp, high end signals passed through an EQ or other signal processor before going to the amplifier). For these systems, a fully balanced input (or isolation transformers) on all channels is the most likely to be able to prevent ground loop noise. In this case, they are used to ensure that each input uses only the shield ground from the signal source as a reference. With other input circuits, the common shield grounds can cause noise from ground loops.
The most common method (used for virtually all budget amplifiers) is to float the secondary of the power supply. This works well as long as all inputs are being driven from the same signal source.
The input circuit that Rockford uses floats the ground for the 15v regulators. This works well but can produce noise unless all signals are from the same source.
Other amps use balanced inputs (or some form of active input circuit). These work best when the inputs of the amp come from different sources (i.e. sub output from head unit directly to the amp, high end signals passed through an EQ or other signal processor before going to the amplifier). For these systems, a fully balanced input (or isolation transformers) on all channels is the most likely to be able to prevent ground loop noise. In this case, they are used to ensure that each input uses only the shield ground from the signal source as a reference. With other input circuits, the common shield grounds can cause noise from ground loops.
Re: TOPAZ
lol lol lol I wasn't worried about RF trying to prosecute an end user adapting there design to suit his one time use and need lol lol catch me if you can in this case lol lol lol
RF amps have a long and very good reputation for noise free installs, Hence my reason for suggesting their approach.
Perry has pretty much covered all methods and there relative disadvantages, and advantages
I myself have found that I have never really needed a noise canceling input on any of my gear.
But in a pinch there are always the high end Jensen studio grade transformers that list at over $110.00 a stereo set.
But using ANY transformer can also create issues as the coils act like magnetic pickups and any placement near a AC generating noise source will possible lead to hum and noise pickup worse then the original problem.
I use these on my test bench to isolate my music and signal source from any device I am testing < I have seen 60 volts DC standing on the RCA shield of some amps like the older Kicker ZX and Zr products. That much juice will surely damage any source you might care to hook up.
The circuit Perry posted should give good results, But nothing is perfect as Perry has warned....
lumanauw said:Rockford has patent #5751823, which it's critical point is choosing to which point the opamp's regulator reference (317 and 337 in fig.5) connected to. It is connected to head unit's ground, separated by R//D to power amp's 12V ground.
lol lol lol I wasn't worried about RF trying to prosecute an end user adapting there design to suit his one time use and need lol lol catch me if you can in this case lol lol lol
RF amps have a long and very good reputation for noise free installs, Hence my reason for suggesting their approach.
Perry has pretty much covered all methods and there relative disadvantages, and advantages
I myself have found that I have never really needed a noise canceling input on any of my gear.
But in a pinch there are always the high end Jensen studio grade transformers that list at over $110.00 a stereo set.
But using ANY transformer can also create issues as the coils act like magnetic pickups and any placement near a AC generating noise source will possible lead to hum and noise pickup worse then the original problem.
I use these on my test bench to isolate my music and signal source from any device I am testing < I have seen 60 volts DC standing on the RCA shield of some amps like the older Kicker ZX and Zr products. That much juice will surely damage any source you might care to hook up.
The circuit Perry posted should give good results, But nothing is perfect as Perry has warned....
What do you guys think about this page http://www.dself.dsl.pipex.com/ampins/balanced/balanced.htm#1
At that page it is different between balanced line and balanced impedance (fig1).
To get balanced impedance, the RCA out ground of the head unit should be given Rs too, but I never saw a car head unit that do this.
At that page it is different between balanced line and balanced impedance (fig1).
To get balanced impedance, the RCA out ground of the head unit should be given Rs too, but I never saw a car head unit that do this.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- General Interest
- Car Audio
- Good Example Balanced Input Circuit? (Ground Loop Isolation)