@Kooka, it seems like there is a problem around Q1. It should be conducting and pulling the output down towards 10V. Check and/or re-solder the connections to Q1. When you measure 0V on the Source of Q1, make sure both the pad on the PCB and the lead of the transistor measure the same Voltage. The solder connection should look good on both sides of the PCB. Occasionally there can be a break in the plating between the top and bottom side of a plated thru hole.
Loudthud, Q1 is very well soldered, on both sides and with very well melted silver solder and all connections to Q1 are well checked and ok...
Do you think the problem could be in the part itself..?In this moment I connected power again, and biasing is set to 8.4V immediately after connecting power. If I try setting it, it stays there dor some trimming degrees and then, all at once, it jumps to 15V. Or, trimming clockwise, it stays to 15V for a few degrees of the trimpot,mand then goes close to 0V...
Last edited:
I can't rule out a problem with Q1, but it would be an unusual problem. Most of the time when these big HexFets go bad, there is a low resistance short form Drain to Source. The Voltage at the bias point would be a few 1/10ths of a Volt all the time.
If Q1 is operating properly you should expect the Gate Voltage to be low, around 1 Volt or so when the Voltage at the bias point is high (+14V), and the Gate Voltage to be high (over 3V or so) when the Voltage at the bias point is low.
If Q1 is operating properly you should expect the Gate Voltage to be low, around 1 Volt or so when the Voltage at the bias point is high (+14V), and the Gate Voltage to be high (over 3V or so) when the Voltage at the bias point is low.
Pass DIY Addict
Joined 2000
Paid Member
When i finish populating a circuit board, one of the things that I always do is to use a fine tip, flat blade screwdriver to scratch between the pads on the pcb. This helps make sure that i did not create a solder bridge, however small, where it doesn't belong. In this case, i would be careful to do this between the pins for Q3 and Q4.
Eric
Eric
Pass DIY Addict
Joined 2000
Paid Member
Kooka, work by process of elimination to solve your problem. Some of these you've already done, but here would be my list of things to check, roughly in order:
1) power supply functioning properly
2) double check ALL parts for proper placement/orientation/value on PCB. Turn the adjust screw on P1
3) are two large transistors electrically isolated from your heat sink? Use a DMM to check for continuity.
4) no cold solder joints on PCB, reflow everything, use a sink on small signal transistor legs to protect device from too much heat
5) no inadvertent solder bridges where they should not be on PCB, use magnifying glass and scratching device
6) replace all four transistors, two small ones first
This is such a simple design (compared to others), the error is likely to be something silly or simple. The solution should present itself shortly. Don't give up!
Eric
1) power supply functioning properly
2) double check ALL parts for proper placement/orientation/value on PCB. Turn the adjust screw on P1
3) are two large transistors electrically isolated from your heat sink? Use a DMM to check for continuity.
4) no cold solder joints on PCB, reflow everything, use a sink on small signal transistor legs to protect device from too much heat
5) no inadvertent solder bridges where they should not be on PCB, use magnifying glass and scratching device
6) replace all four transistors, two small ones first
This is such a simple design (compared to others), the error is likely to be something silly or simple. The solution should present itself shortly. Don't give up!
Eric
Last edited:
Thanks guys, I appreciate!
I think I already did everything you suggest, but I will take some free time before facing again the thing, just to clean up my mind and do it agan.
Will keep you posted.
Thanks guys, I appreciate!
I think I already did everything you suggest, but I will take some free time before facing again the thing, just to clean up my mind and do it agan.
Will keep you posted.
No need to power on, check Speaker binding (output) resistance, see if any shorted
Dear Fourm ... been lurking a while, and have my ACA boards + parts together.
A question re Solder ... In my other hobby, with Raspberry PI I use proto boards and have built a few. For this, I use a paste solder, and its been working just fine.
I find it much easier to control than wire solder. Never had issues with cold joints etc ...
The various searches that I have done on the forum turn up nothing. Any thoughts?
Thanks
Dave
A question re Solder ... In my other hobby, with Raspberry PI I use proto boards and have built a few. For this, I use a paste solder, and its been working just fine.
I find it much easier to control than wire solder. Never had issues with cold joints etc ...
The various searches that I have done on the forum turn up nothing. Any thoughts?
Thanks
Dave
Thanks guys, I appreciate!
I think I already did everything you suggest, but I will take some free time before facing again the thing, just to clean up my mind and do it agan.
Will keep you posted.
I the past when in situations like yours I turn off power and disconnect all inputs and outputs and grab my digital meter. You have the advantage of two channels that you cAn compare side by side. Make sure that the bias pot of the bad channel is set to the same rotation as the good channel. Connect the negative ohmmeter lead to a jumper between the grounds of both channels and check first a point on the working channel And then the corresponding point on the non working one. The various capacitors in the circuit will cause readings to vary as they charge. Wait until you get a stable reading and then record the result. You can change the pot settings of the channels to other equal settings (full clockwise and full counter clockwise for example) and repeat the checks. If your meter has a diode Check ohms range you can use that meter range to apply potential to forward bias some of the semiconductor junctions in the circuit.
The point of this exercise is to compare the two circuits in a safe way looking for a difference that you can then analyze to hopefully find the problem. One other idea is to get some other eyes looking at the circuits. Perhaps if you could take some photos and post here someone may see something that you have missed.
Last edited by a moderator:
Thanks guys, I appreciate!
I think I already did everything you suggest, but I will take some free time before facing again the thing, just to clean up my mind and do it agan.
Will keep you posted.
I the past when in situations like yours I turn off power and disconnect all inputs and outputs and grab my digital meter. You have the advantage of two channels that you cAn compare side by side. Make sure that the bias pot of the bad channel is set to the same rotation as the good channel. Connect the negative meter lead to a jumper between the grounds of both channels and check first a point on the working channel And then the corresponding point on the non working one. The various capacitors in the circuit will cause readings to vary as they charge. Wait until you get a stable reading and then record the result. You can change the pot settings of the channels to other equal settings (full clockwise and full counter clockwise for example) and repeat the checks. If your meter has a diode Check ohms range you can use that meter range to apply potential to forward bias some of the semiconductor junctions in the circuit.
The point of this exercise is to compare the two circuits in a safe way looking for a difference that you can then analyze to hopefully find the problem. One other idea is to get some other eyes looking at the circuits. Perhaps if you could take some photos and post here someone may see something that you have missed.
Does the paste solder contain flux? If it is water soluable flux, it is acid based and it will need to be removed completely. In production, a dishwasher is typically used but I don't know if they use straight water or some kind of soap and abrasive "helper".
The solder specs ..
This is a low temp solder paste with a no clean flux
Melting Point 137C 278F (Eutectic)
87% Metal Particle Size 25-45 Microns(T3)
Protect Sensitive Parts with this low melting Lead Free Solder Sn42/Bi57/Ag1
This alloy is better than Sn42/Bi58 because it contains Ag which makes it stronger and less brittle
It does contain a flux, and while they say no clean, I usually use an old tooth brush and a bit of alcohol. everything seems nice and shiny.
Temp seems good as the caps in the amp are 105 degree as I recall.
Dave
This is a low temp solder paste with a no clean flux
Melting Point 137C 278F (Eutectic)
87% Metal Particle Size 25-45 Microns(T3)
Protect Sensitive Parts with this low melting Lead Free Solder Sn42/Bi57/Ag1
This alloy is better than Sn42/Bi58 because it contains Ag which makes it stronger and less brittle
It does contain a flux, and while they say no clean, I usually use an old tooth brush and a bit of alcohol. everything seems nice and shiny.
Temp seems good as the caps in the amp are 105 degree as I recall.
Dave
