The amp settled in at about 1.7A bias last night with the LED's in place of the 1N4001's. After being on for about two hours, the both the regulator and heatsinks measure about the same temperature of 59 C (regulator sinks were hovering around 70 C before swapping in the LEDs). I'll do some temperature testing with the lid on vs off.
Some measurements from tonight. The ambient temperature of the room was 26 C.
I first powered on the amp without the lid, and let it stabilize for 2 hours. It measured:
T heatinks = 59C
T reg heatsinks = 60 C
Amp Bias = 1.66A
I then placed the lid on the amp and let it sit for another 1 1/2 hours. It then measured:
T heatsinks = 63 C
T reg heatsinks = 72 C
T lid = 57 C
Bias = 1.75A
So the bias drifted up about 0.1A and the regulator heatsinks went up ~10 to 12 C in temperature. The regulator mosfets were dropping between 2.85V to 3.05V, depending upon which one you measured.
I think I will turn down the bias so it stabilizes between 1.5A and 1.6A with the lid on, and be done with it.
I first powered on the amp without the lid, and let it stabilize for 2 hours. It measured:
T heatinks = 59C
T reg heatsinks = 60 C
Amp Bias = 1.66A
I then placed the lid on the amp and let it sit for another 1 1/2 hours. It then measured:
T heatsinks = 63 C
T reg heatsinks = 72 C
T lid = 57 C
Bias = 1.75A
So the bias drifted up about 0.1A and the regulator heatsinks went up ~10 to 12 C in temperature. The regulator mosfets were dropping between 2.85V to 3.05V, depending upon which one you measured.
I think I will turn down the bias so it stabilizes between 1.5A and 1.6A with the lid on, and be done with it.
Last edited:
Yes, I was also told by others that 1.5A bias results in less drift after 1 hour.
BUT you have to bear in mind that I did the design for 20°C room temperature.
Then the regulator sink would be at 66°C with the lid on, and that is totally OK.
For summer use in e.g. Spain, you may consider adding water cooling.
It is quite easy and works like magic.
You only need some silicon tubes (6mm outside diameter) to connect to a big water tank on your balcony.
Some small length of copper tubing and plates, and an aquarium pump.
Patrick
BUT you have to bear in mind that I did the design for 20°C room temperature.
Then the regulator sink would be at 66°C with the lid on, and that is totally OK.
For summer use in e.g. Spain, you may consider adding water cooling.
It is quite easy and works like magic.
You only need some silicon tubes (6mm outside diameter) to connect to a big water tank on your balcony.
Some small length of copper tubing and plates, and an aquarium pump.
Patrick
It is quite easy and works like magic.
You only need some silicon tubes (6mm outside diameter) to connect to a big water tank on your balcony.
Some small length of copper tubing and plates, and an aquarium pump.
Patrick
Why did I suddenly get The Red Green Show in my head when I read this?
https://www.youtube.com/watch?v=MpJM4Y1ajew#t=325
So just recently, the left channel on my amp is having issues. After about 10 minutes of playing time, a loud buzz develops for about 5-10 seconds and then the left channel cuts out. I can't tell exactly, but it sounds like it might be coming from either the left channel transformer or the board itself. The bias of the left channel remains exactly the same, the left channel regulator board remains on, and the voltage in the left power supply does not change whatsoever.
The right channel keeps playing as normal, and only the left channel muting board seems to turn off. I've been able to replicate this issue 3 times tonight. Any ideas what the problem might be?
Thanks
The right channel keeps playing as normal, and only the left channel muting board seems to turn off. I've been able to replicate this issue 3 times tonight. Any ideas what the problem might be?
Thanks
You are an experienced DIYer.
I am surprised you did not do more diagnosis before posting.
I have also described the functioning of the protection board in detail.
The protection relays all have one trip signal line and are not separate left and right.
So if one channel has a DC (fault), both channels will be turned off.
If only one channel turns off, then your relay driver for that channel is not working properly.
If I remember correctly, you put 24V on 12V relays without adding compemsating resistor.
So you should look at that.
You should also look at the gate voltage of the buzzing relay.
Maybe you have oscillation there.
The best way to find out whether the amp itself has a problem is :
1) Take the amplifier module off and use a lab supply with current limit.
2) Short the inputs and check all DC values. You should attach a 4~8ohm resistive load.
3) Use a functions generation to check output under load with an oscilloscope.
4) Repeat (3) with real speaker load.
If all OK, then the problem is not with the amplifier.
I have life and death issues to deal with at home.
So I shall not be able to nurse you through this for at least the next 2~3 months.
Patrick
I am surprised you did not do more diagnosis before posting.
I have also described the functioning of the protection board in detail.
The protection relays all have one trip signal line and are not separate left and right.
So if one channel has a DC (fault), both channels will be turned off.
If only one channel turns off, then your relay driver for that channel is not working properly.
If I remember correctly, you put 24V on 12V relays without adding compemsating resistor.
So you should look at that.
You should also look at the gate voltage of the buzzing relay.
Maybe you have oscillation there.
The best way to find out whether the amp itself has a problem is :
1) Take the amplifier module off and use a lab supply with current limit.
2) Short the inputs and check all DC values. You should attach a 4~8ohm resistive load.
3) Use a functions generation to check output under load with an oscilloscope.
4) Repeat (3) with real speaker load.
If all OK, then the problem is not with the amplifier.
I have life and death issues to deal with at home.
So I shall not be able to nurse you through this for at least the next 2~3 months.
Patrick
Last edited:
Thanks Patrick.
As I thought about it more late last night, it did seem like the problem was at the muting board. I never knew a small relay could buzz so loudly. I will investigate more, and try swapping out the relay for a new one.
I'm sorry to hear about your issues at home. I wish you and your family the best of luck through this time.
As I thought about it more late last night, it did seem like the problem was at the muting board. I never knew a small relay could buzz so loudly. I will investigate more, and try swapping out the relay for a new one.
I'm sorry to hear about your issues at home. I wish you and your family the best of luck through this time.
When the buzzing problem started last night, I took voltage measured across the relay. I measured 24V across the diode on the relay board prior to the buzzing, and then 0V afterward so I am thinking the relay itself may be ok. On this amp, I swapped out the input relays for 24Vdc ones before I powered it up for the first time.
I am starting to think it may be an issue with the input relay driver on the protection board. I will do some more testing tonight to confirm it is not the relay itself, and see if I can better isolate the problem.
I am starting to think it may be an issue with the input relay driver on the protection board. I will do some more testing tonight to confirm it is not the relay itself, and see if I can better isolate the problem.
Greg,
You would find it a much more rewarding experience if you would take your time and eventually figure out and cure the problem yourself.
Then it would be the right time to post your experience here.
No one else can do remote debugging for you, and more importantly, you don't need that.
Patrick
You would find it a much more rewarding experience if you would take your time and eventually figure out and cure the problem yourself.
Then it would be the right time to post your experience here.
No one else can do remote debugging for you, and more importantly, you don't need that.
Patrick
