Hey all,
i had some probelms with my 7 channel gainclone amp. i had that irritaing buzz in all the 7 channels.
I then replaced all input wires with shielded cables.
The result: the humm reduced to almost imperceptible levels. However, when you turn it on say after 11 in the night, i can make out a very very faint buzz.
Couild groudning be the cause of this.
I have three EI type transformers in my amplifer case.
one is 650VA for the bridged 4780s and OPA541.
The other is 350 VA for the paralleled Lm4780 and a pair of islolated LM3886.
The final one is for the 3 way active crossover.
Now, i have taken one wire from each of the three capacitor bank grounds to one point in the middle of the chassis.
is this the right way to go about it?
What is the normal procedure that you follow while routing ground tracks inside ur amp?
I need some help here guys.
thanks.
Sachi
i had some probelms with my 7 channel gainclone amp. i had that irritaing buzz in all the 7 channels.
I then replaced all input wires with shielded cables.
The result: the humm reduced to almost imperceptible levels. However, when you turn it on say after 11 in the night, i can make out a very very faint buzz.
Couild groudning be the cause of this.
I have three EI type transformers in my amplifer case.
one is 650VA for the bridged 4780s and OPA541.
The other is 350 VA for the paralleled Lm4780 and a pair of islolated LM3886.
The final one is for the 3 way active crossover.
Now, i have taken one wire from each of the three capacitor bank grounds to one point in the middle of the chassis.
is this the right way to go about it?
What is the normal procedure that you follow while routing ground tracks inside ur amp?
I need some help here guys.
thanks.
Sachi
Hi
In all honesty, it sounds like you have an absolute dynamite as far as your setup is concerned, and I don't mean that as a compliment.
First, you have housed the crossover and power amp in the same case, and then are feeding them off separate supplies. I would imagine a single massive 1000VA transformer would be far more suited for such a complex arrangement, with a single supply regulated down to the voltage required for the crossovers.
Having said that, there are a few questions.
1. Once you disconnect the input, is the hum still there?
2. Is the hum also present when the power to the transformer is disconnected (but the caps still have a bit of charge - you can detect this in the first 1 second of switch-off).
3. Where have you returned the speaker and signal grounds to? It's not just about power grounds.
In all honesty, it sounds like you have an absolute dynamite as far as your setup is concerned, and I don't mean that as a compliment.
First, you have housed the crossover and power amp in the same case, and then are feeding them off separate supplies. I would imagine a single massive 1000VA transformer would be far more suited for such a complex arrangement, with a single supply regulated down to the voltage required for the crossovers.
Having said that, there are a few questions.
1. Once you disconnect the input, is the hum still there?
2. Is the hum also present when the power to the transformer is disconnected (but the caps still have a bit of charge - you can detect this in the first 1 second of switch-off).
3. Where have you returned the speaker and signal grounds to? It's not just about power grounds.
sangram said:
well, the reason for going for two seperate transforemrs for the amps are that i wanted to keep power dissipation under contrl.
sO i chose my secondaries according to the cip/config they were going to be used in.
Also, a 1KVA EI would be insanely big.
i could have eliminated the crossover trannny ..had originally thought of making a seperate case for it but then decided to put it all in one case. by then i had already ordered my transformers.
hence went in for a thrid.
1> I will check this and let you know this evening.
2>i donot have soft turn on/off circuitry and DC protection circuitry yet!. so i get a loud pop from the bass driver while the mids and the highs make a soft clicking sound.
will make a soft turn on/off and DC protection ckt soon. one each for the mids and the highs.
3>The input RCA connector's ground is directly connected to the chassis(no insulation between the chassis and the ground pin)
The speaker output is grounded on the amplifeir boards to the power supply grounds.
The star ground should not be connected to the chassis directly, that is the point.
The star ground is a physical point. The chassis is a physical plane, with various voltage drops occuring across its resistance depending on distance. The hum you hear is the voltage drop between the entry point where the plug is physically connected, and the star ground, plus amplification.
So the star ground needs to be lifted off the chassis via a small resistor, and the signal, power and output grounds returned to the star ground. The chassis is normally connected to earth terminal (household electrical earth).
The way you have done it is to return all power grounds together. This so far is correct. In addition you should return all output grounds (speaker terminals) and input grounds (input signal) to the ground at the same physical point. Then this point should be connected to the chassis through a 5-10 ohm resistor (usually 2 watts), and not directly like you have probably done.
The star ground is a physical point. The chassis is a physical plane, with various voltage drops occuring across its resistance depending on distance. The hum you hear is the voltage drop between the entry point where the plug is physically connected, and the star ground, plus amplification.
So the star ground needs to be lifted off the chassis via a small resistor, and the signal, power and output grounds returned to the star ground. The chassis is normally connected to earth terminal (household electrical earth).
The way you have done it is to return all power grounds together. This so far is correct. In addition you should return all output grounds (speaker terminals) and input grounds (input signal) to the ground at the same physical point. Then this point should be connected to the chassis through a 5-10 ohm resistor (usually 2 watts), and not directly like you have probably done.
First thing to rememberis that there is no perfect conductor, any peice of wire will have some resistance... so if you had 3 points on a wire 2 of which were closer to each other, than to the 3rd, there would be less resistance between these two points than between either of those two and the 3rd point... its a very small amount but its still there.
Now consider that all the "noise" is dumped to the ground, and you want to connect your input to that dirty ground... not all the noise going to the ground will end up there if it finds paths of less resistance to say for instance your input grounds, and so contaminating your clean incoming signal... using the thin wire or pcb trace to connect your signal ground makes it so that the noise stays on the ground rather than take the hard road back to your input...
This is very generalised, but if you could visualise this, you know all you need to.
Also what my friend has said above is all valid, excep I would like to add you want to add a 0.1uf cap in parallel with the 10R resitor, and in parall with those 2 big diodes placed so that the cathode of one connects the anode of the other...
Now consider that all the "noise" is dumped to the ground, and you want to connect your input to that dirty ground... not all the noise going to the ground will end up there if it finds paths of less resistance to say for instance your input grounds, and so contaminating your clean incoming signal... using the thin wire or pcb trace to connect your signal ground makes it so that the noise stays on the ground rather than take the hard road back to your input...
This is very generalised, but if you could visualise this, you know all you need to.
Also what my friend has said above is all valid, excep I would like to add you want to add a 0.1uf cap in parallel with the 10R resitor, and in parall with those 2 big diodes placed so that the cathode of one connects the anode of the other...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.