DH-220 left channel dead

[rfarn]

Just picked up a Hafler DH-220. Guy said it was working. Left channel is actualy dead. All fuses are fine. I suspect the FETs aren't powering up. The left heat sink does not heat up. Nothing seems to be burned on the board. All connections look good. Power is getting to both the + and - sides of the board. The sound is find on the right channel. Left channel is just an occasional garble. What should I check?

[Leolabs]

Start from input socket and wirings.

[anatech]

Hi rfarn,
Have a look and make sure all the wires are connected to the PCB first. The cold heatsink means there is no bias current.

How much amplifier servicing experience do you have? There are some common things that should be serviced even if the amp is working.

-Chris

[mmerig]

Here are 3 quick checks to get started:

1: Swap the MOSFETS from one channel to the other; pay attention to polarity. This is easier than unsoldering and swapping boards. If the MOSFETS are okay . . .

2. Check Q9, which is part of the bias circuit.

3. Check C1, especially if it is a bigger, heavier replacement. It's possible that a bad solder joint jiggled loose during handling.

If none of this helps, search the site for Hafler problems. Do you have the manual?

[rfarn]

Thanks for the replies. Chris, I have little servicing experience, but I do have a Physics degree and I want to understand more of how this works. I can replace components easily and would like to do as much as possible on my own.

Excuse the ignorant questions...What is the best way to check Q9? What is the bias curcuit/current?

mmerig, I'll check that when I get home tonight.

Any other suggestions would be greatly appreciated



Bob

[anatech]

Hi Bob,
1) Never swap components from one channel to the other to troubleshoot. It's a great way to end up with two damaged channels.

2) Check bias currents and DC offsets first, then check gate voltages with respect to the output terminal.

3) Study the schematic. Your head is the best tool you have.

4) If you are lucky, you may have some intermittent solder joints. An intermittent bias control is possible.

If you have a variac, use it to power up each time whle watching your current draw.

What you may want to do is purchase all the new capacitors on the board and replace them. Do the bad channel first, do not touch the working one until the other one is working well.

Take digital pictures of the board before you start.

-Chris

[mmerig]

As usual, Chris (anatech) is right .. it is risky to swap components. I got the idea of swapping the MOSFETS from this site, and it helped me trouble-shoot my amp, but I had checked voltages at certain places first to avoid potential problems.

I made a MOSFET checker later too, (for matching), and this is a safer way to check them. In my case, I thought the MOSFETS were bad, but they were okay. Feed back on this site and other places suggests that the MOSFETS are probably okay. They are tough devices.
Nelson Pass has an article on testing MOSFETS:
http://passdiy.com/pdf/mos.pdf

The transistors on the board are easier to check. Here is a link on one way to do it:
http://ourworld.compuserve.com/homep...tt/elect56.htm

I had a muffled channel on a Hafler 220 and I never found the problem. After lots of checking, some mistakes on my part, and good advice from this dyi site, I finally cheated and bought a used board for 5 dollars or so from John Hillig at Musical Concepts. I may try and fix that old board for fun, but from a time and cost perspective, the used board was a practical way to get my amp working again.

Are there "new capacitors" on the board (re anatech's message)? Replacing or at least checking them is worthwhile. I think replacing all of the capacitors will be costly and time consuming. Many of them are film caps. From what I've read (and some experience) film caps don't fail very often. It would be worth replacing the electrolytic caps, as they tend to wear/change/degrade with age. There are only three electrolytics; C8 is in the signal path and a more likely source of the problem.

Here is alink on capacitor testing, etc.:
http://www.repairfaq.org/sam/captest.htm

[anatech]

Ahhh mmerig,

There are only three electrolytics; C8 is in the signal path and a more likely source of the problem.

Got me there!
That is of course what was the intention. Most of them I see are bad. The series resistor for the supply decoupling cap may open up as well.

You should try to repair your board for the experience. While you are at it, match the two differential pairs per board.

You may want to use a modern DVM on diode check to test the small transistors. The method shown in the link above is rather hard on the base emitter junction. Most new meters also have an hFE test. Use it to both test and match your signal transistors.

You younger guys have things so easy!

-Chris

[fab]

You may check also D7, D8, D9, D10 and b-e junction of Q12 and Q13 if they are shorted if you have no bias.