DH-220 DC Offset, to adjust or not.

pkquat

I have an all original DH-220 minus new RCA inputs that I recently put back in service. I checked the DC offset recently and it is approaching 30mv for each channel. It used to be around 20mv as I recall. Do I dare try adjusting the DC offset pots or leave them until I have replacements on hand?

tonyb

I don't see why you can't adjust the DC offset. Is this the amp in question ?

http://www.acid-house.com/toys/audio/hafler-dh-220-power-amplifier

If so, it's not a very powerful amp, not even rated for 4 ohms. You sure you want to put the work into it ?

pkquat

I've heard with older amps that due to the age and oxidation build up, the pots can be a nightmare to adjust and don't clean up well. It was often easier to replace them than trying to dial them in than trying dial them in. I am not sure if that is true for this amp. I'm hoping someone has some experience.

Yes that is the amp in the link. It is 4 ohm rated and can do less. In the 90's it ran some ribbons and hard to drive speakers rather well and close to their limit. I haven't done solid A /B testing, but it keeps very good pace if not even pace with a 150W @ 8 ohm Krell 300i. The Hafler is not a power house, but not bad on a budget. I have yet to own speakers that drain the amp.

heiney9

pkquat wrote: »

I've heard with older amps that due to the age and oxidation build up, the pots can be a nightmare to adjust and don't clean up well. It was often easier to replace them than trying to dial them in than trying dial them in. I am not sure if that is true for this amp. I'm hoping someone has some experience.

Check the DIY Audio website, way more techy knowledge over there than here. I think you've answered your own question. Best to have the new pots on hand and even better to install them.

H9

pkquat

TY I thought I would check here first since I've been hanging around here lately and there seem to be some knowledgeable people around. I am not an electrical wizard and don't know components or manufacturers too well, so I was hoping to hear the risk of adjusting them would be minimal or maybe find someone who knew what to get. Until I found this forum, I had no idea of what brands to get or that it is recommended to replace resistors when rebuilding crossovers.

heiney9

Amps are whole different breed compared to x-overs, so I'd proceed with caution. Both because of danger of electrocution or really messing something up. Do you have the service manual or are you just sort of guessing?

Very hard to do any repairs on an amp without the service manual.

H9

pkquat

No worries there. The Hafler manuals are online and I have copies. I have some knowledge, and skills with electronics, AC power and not getting zapped, just not the fine points of components, how much to spend for what, or circuits of the top of my head. Adjusting the bias and dc offset with the pots are pretty straight forward for me (they are boldly labeled on the Hafler boards). Its more diagnosing something if it goes wrong. What I meant was, I know I might be able to find a pot at Radio Shack or online that in theory would match, but it might be a downgrade, or create more problems, i.e. sandcast resistors vs. Mills, or a wide tolerance cap, etc. Or I buy the one for $50 when the $5 would do fine, but the $3 one is junk.

tonyb

Try adjusting it first, cost you nothing.

K_M

http://audiokarma.org/forums/index.php?threads/amplifier-distortion-dc-offset-and-you.5634/

heiney9

To reiterate the warning at the bottom of the above link:

Edit:***WARNING!!!***

Since the start of this thread, which was simply intended to have people LOOK at offset, many have decided that they want to try and ADJUST it.

***DO NOT TRY TO ADJUST OFFSET OR ANY OTHER DAMN THING INSIDE YOUR AMP OR RECEIVER UNLESS YOU REALLY REALLY REALLY KNOW WHAT YOU ARE DOING!!!***

I can forward you emails and AK thread links concerning guys who have destroyed their amps/receivers because they thought that all that they has to do was 'a little tweak'. Instead, they cranked up the bias pot instead of the offset pot, either because their amp/receiver HAS no offset adjust (many don't), or because they picked the wrong pot to tweak. OR, a slip of the probe lead and POOF! goes the output transistors and many other parts. So read and re-read the ***ALL CAPS*** statement above until you fully comprehend it.

...and buy the damn service manual! :yes:

pkquat

Thanks. The pots are clearly labeled, not near each other, the values are covered in the manual, and there a lot of space. One of the benefits of this Hafler is it was available as a kit, so everything is clearly labeled, straight forward, and well spaced. From my preliminary poking around online, I should be fine with these pots, but I want to find out what the replacement ones are first, so if I need to, ordering them will be easy.

From what I have gathered the issues were more often with the small square vertical pots, amps that had a bad bias and were running hot for a while, and very dusty and dirty ones. The other issue, mainly while trying to adjust the bias, is that the bias takes some time to settle. Any movement triggers a wild swing before it settles, which can make people think the pot is bad.

ambiophonics

I usually give those pots a little squirt of cleaner with the amp off first and rotate them back and forth making sure to put them back in the same position they started. Then I power the unit on, let it settle and adjust the DC offset and bias as outlined in the manual. Some are more difficult to dial in than others but I don't usually find the need to replace them after cleaning. I prefer deoxit D5 for the cleaning solution.

pkquat

Good to know. Thanks. I should getting to it this weekend.

pkquat

All went well as far as DC offset. ±1mv now. The pots were in very good shape. Although my amp is quite clean, I did give them a shot of Radio Shack dry de-oxidizer, then some D5 and let that sit for a couple days (I meant to let it sit for a couple of hours, but plans changed). The values settled quickly after peaks of 20-60mv. It's my understanding that adjustments should be done in small increments to minimize spikes, even after you know which direction to turn the pot and and take a stab at where it needs to be.

pkquat wrote: »

The other issue, mainly while trying to adjust the bias, is that the bias takes some time to settle. Any movement triggers a wild swing before it settles, which can make people think the pot is bad.

Correction The DC offset takes a wild swing and then settles, not the bias, at least in my case. The bias was pretty uniform. The DC jumps then settles. This is mentioned in the manual.

I recommend not doing anything with the bias if both sides are the same temp and only running warm, unless you are familiar with adjusting it. If one or both sides are hot, closer to Krell temps, then the bias may be too high. Not very warm, and they are too low. Any adjustment to the bias will effect the DC offset. DC offset needs to be adjusted last.

I did run into an issue checking the bias, and need to dig further into that process and concerns. Tip, do use clips to maintain contact, don't probe. Both sides had been at the same temp and about the right temp. I followed the process in the manual and one side was very close to 275mA, after the turn on spike, although it slowly climbs. On the other side, while I had one part clipped and was holding the probe, I broke contact, and re-established it and the reading jumped off scale (>320 mA), and then I broke it again before turning the amp off. I waited for a while, but maybe not long enough before making contact again, it was still high, over 400 mA. I did smell a little ozone and that unhappy electronics smell. I waited long while, and checked it again. It was around 400mA. This is out of spec, but not out of range. I decided to tackle it later when I had two clips because now I needed to adjust it.

I put the fuse back in the next day. I powered it up to let it warm and the one side was running hotter. I powered it off waited and then checked the mA again, this time with clips. The good side was still fine. The bad side was about 380mA to start and then climbed. I was able to adjust the bias without much difficulty. I wound up turning the pot a only few degrees. Since the bias seams to climb while it is on, I powered it down waited for the mA to drop to around .2 and then turned it on again. It was too low. Eventually I wound up getting the pot very close to its original location. Now both sides are about 275mA after the initial spike. After putting the fuses back in, both sides are the temps they were before. After this adjusting the DC offset was easy.

I'll ask over at the DIYaudio forum about the bias unless someone here knows. The bias always creeped up to around 285mA after 15-20 seconds and to around 295 after a minute. It was still climbing, but very slowly. I am not sure if this is normal or not. I am also curious what the negatives were when I broke contact and re-established it. I assume I need to let everything dissipate between readings. I noticed that when I disconnected the clips and closed the contacts with the fuse for a few seconds before hooking the clips back on and powering on the amp, I was right on spec. after the spike. If I let it drop to 0.1-0.2mA, without disconnecting and closing the contacts the initial current was higher (+5-10mA ) when I powered the amp on again. This part exceeds my electrical knowledge and was not in the manual.

pkquat

One more update. For all adjustments the amp needs to be left on for a significant amount of time to allow things to settle. The bias settles within a few minutes, but the DC offset can take much longer to really settle in. Everything needs to be done with the amp warm, and preferably covered. The values changed some once the cover was in place. The bias can be greater than 275mA, but should not really be over 300mA. The consensus is that it will run too hot. 125°F should be about the max temp, and both sides should be within a couple of degrees.

The big note is since it is a common ground amplifier changing the DC offset on one channel effects the offset on the other. This drove me NUTS. I am sure their are tricks, but it took me a while. I got them down to -7mV each. The readings are fairly stable, but I noticed over time, possibly temperature, they can vary by a few mV in either direction.