Hafler XL-280 might need transformer? Please help!

[HaflerFreak]

Hi you guys, this is my first post. There is a very helpful post with similar symptoms here:
help repairing Hafler xl280

My problem is this:
In the past I have used my XL280 to amplify a variable low-level output from my TV. I connect the TV's output to an input on my DH110 preamp and set the volume on the preamp at about 1/3, I can then control the volume with my TV remote.

Recently, while watching something interesting, I turned the volume up near max on the TV. One side of the amp cut out to low-level distorted sound. I immediately turned the amp off. There was no indication of thermal overload (both channels cut out and the power switch lamp blinks). Later, I turned the amp back on with a regular CD source through my preamp. Both channels appeared to work fine at first, but upon critical listening I noticed that the side that had cut out was now producing a 'tinny' nearly bassless output. Other than the lack of bass, it is not noticeably distorted in any way. I have tried switching input wires and speakers - it is the amp.

I pulled the cover and looked for anything obvious - nothing.

As a matter of course due to its age, I replaced the 100V 470uF caps and all 4 1N4003 diodes on the power supply board. Put everything back together and measured the voltages. I get a solid +-69 volts on both sides of what should be the +-75 volt supply and +-60 volts from what should be the 65 volt FET voltage.

Keep in mind one channel sounds fine with strong bass but they both show low voltage from the supply. Also, the ratio of voltages appears near correct (60/69 compared to 65/75) just 5 or 6 volts low.

I checked the bias but I need to get a multimeter that accurately reads up to 400 mA. With the meter I have bias reads .3 A and it should be 300 mA so it does not appear be too far off. Both sides of the amp appear to be pretty evenly warm after 15 minutes. Could low bias affect my voltage readings?

I think I need a new trasformer but I have not yet done the specific tests to find out.

Does anyone have any testing suggestions or experience/info that they think would be of help? I love this amp and I will buy a second one to gut if I have to.

Anyone have a good, used XL280 transformer to sell?

[gearheaddruid]

Hi, there is not a problem with the transformer, your incoming line voltage is a little low and that would explain the rails being off by about 5 volts.
What you are describing sounds like a dead capacitor in the feedback circuit or the input. If the channel is tinny and has lower gain than the other, I would suspect the feedback circuit.
Regards, Steve

[HaflerFreak]

Thanks Steve! Wow, this forum really is amazing...

That was exactly what I wanted to hear, caps are easily replaced (with the exception of the 7800uF jobs).

One of the first things that I suspected was AC line voltage when I made the tests - it was about 116~118 volts when I took my readings so you were right on that.

You can probably tell that I am a bit new to this, so please excuse any dumb questions...

What you are describing sounds like a dead capacitor in the feedback circuit or the input. If the channel is tinny and has lower gain than the other, I would suspect the feedback circuit.

So, at the risk of sounding like someone who shouldn't be taking nice amplifiers apart -
How would I identify the feedback circuit or input caps?

In addition to the 100V 470uF power rail caps there are two 100V 100uF 'lytics (I assume for power supply smoothing) and a 6.3V 1000uF Bi-Polar on each driver board. Do you think it is one of the electrolytics?
Here is some great info from another post concerning the bipolar cap:

The Xl-280, starting from the speaker terminal, has a 47k resistor in series to the negative input of the differential pair. In parallel with this resistor is a 3.3k resistor in series with a 10pf capacitor to compensate the feedback loop.

From the differential pair input there is a 270 ohm resistor in series with two parallel high frequency compensation capacitors in parallel to ground. One is 330pf fixed, and the other 120pf variable. That 120pf capacitor is the excilinear (null) adjustment.

There is also a 2.2k resistor in series with a 1000uf NP electrolytic in parallel with a 0.1uf poly film capacitor to ground to set the gain and low frequency cut off.

Would you suggest that I test the 1000uF 'lytic first or the .1 poly film?

BTW here is the schematic link:
http://www.hafler.com/techsupport/pd...80_amp_man.pdf

[gearheaddruid]

Hi again, I would suspect the 1000ufd electrolytic. If it has changed value, it will kill low frequency response and reduce gain. I have seen this problem on a lot of other amps, so that would be my first suspect. The 2.2k resistor may be open also . I could not guess as to why it would have failed. It may just be coincedence with you connecting it to your video system or there was an inaudible high frequency oscillation that damaged these parts. Swap the caps between channels and see if the problem moves with the cap. Regards, Steve

[HaflerFreak]

Hello to Steve and anyone else reading this thread -

I didn't want to leave you hanging, after beginning my new project, I happened upon an interesting story.

First, some history, I bought this particular XL280 'used' from a local HiFi shop about 18 years ago. The dealer said it had been traded in with a problem and that it had been repaired. Needless to say, it was discounted and I got a sweet deal with warranty at $525. Sounded just beautiful with my Magnepan MG2.6/R's.

Flash forward to the present, amp has had 2 or 3 (at the most) thermal shutdowns in the past, once due to my dumba** stacking a component directly on top of it after a move. Always came back until after I modded it. Leading me to my confession - I suspected the BiPolar cap because it was one of the caps I recently replaced while doing an electrolytic 'freshening'. I was a little wary of it because of all the (normally perfect solder jobs I did, I remembered holding the tip just a little too long on that one.

So... So I think you were right Steve! I pulled that 1000uf (It had a 2004 date code BTW - stupid Mouser) and I think I fried it. My ohmeter read all over the place - but it never went infinity. Ohhh yeah, this may be too easy.

THE MYSTERY:
So this is why I put down the history above...
I had pulled this baby apart a couple times, but until now there was something I never noticed.

C1 on both sides had been replaced with a 47pF 5% Mica!!!

That is right smack dab in the signal path, literally the first thing the signal 'sees' and it is supposed to be a 47pF 3% Polypropylene.
Wow. All this time. 18 years with a pair of Maggies and I have been listening to them through as much as 5% Dielectric Absorption. Dear God.

I have ordered some 47pF Polystyrene caps off of eBay - I am expecting a religious experience when this is through.

Thank you DIYaudio, in my gratitude I am posting pics of my progress and I will report anything I 'hear' - and yes, I have golden ears.

Note the XICON Low ESR 100V100uF caps.
Also, check out that polyester at C2 - I've already got a WIMA polypropylene ordered for that.
I think I may upgrade all the polycarbonate caps to propylenes -- anyone have any reasons I shouldn't do that?

[HaflerFreak]

Hello everybody,

I must say, this is a very informative site. I have been reading everything I can find about capacitors lately, and I have learned something very important - not to fix what ain't broke!

The 2.2k resistor may be open also . I could not guess as to why it would have failed.

Ok, I tested R26 and it was right on. I pulled the polycarbonate in the NFB circuit and it tested a little low, but Ok (I think). I replaced it with a WIMA MKP4 met. polypropylene that tested within 2% of spec.
I'm still hoping it was my BiPolar electrolytic that caused the channel to cut out - I didn't bother to switch the cap to the other channel to test it, I just got new ones from Mouser. The new caps are acutally a lot fresher than the first replacements, these are from November of last year. Should I still reform them?

Notes on pictures:
The thing is, I had recently experimented with some Sprague 715P orange drops at C5 on each driver board (see previous pictures), upon further study I realized that it was probably not a good thing to mismatch C5 and C6. Not only that, I found enough disparaging info about the 715P's to dissuade me from using them at all.
The original caps at C5, C6, and C3 that I found in this amp were SEACOR PFDR series 630V .001uF. The schematic calls for .001uF 250V 10% polypropylenes. I ordered some XICON 1431 series 630V .001uF and they turned out to be a precise physical match to the original SEACORs. I measured out 2 'cherries' that tested within a couple picos of each other and installed them.

From reading the Walt Jung/Richard Marsh article 'Picking Capacitors pt1/2' I learned specifically what caps I might want to change as opposed to leave alone.
My reason for changing a few caps here and there is to improve the quality where I can - not to mention extend the life of the amp. After reading other posts on this site, I realized that a commercial design like this doesn't need much 'tweaking' - especially if I have enjoyed it for years. I decided to limit my mods to obvious upgrades (like the MICA to Polystyrene switch at C1), repairs, and basic same type/same value replacements.

On to the capacitor questions:
1. I was considering replacing the 1KV .01uF ceramic across the transformer AC with a 1KV .01uF HV polypropylene (DPPM series) see pics. Would this be good?

2. Does anyone have an opinion about SEACOR vs. XICON vs. 715P Orange Drops?

3. I decided to leave the 330nF polyester at C2 because the WIMA polypropylenes were 15 nF low and the original was within 1 nF of spec. Would anyone still go with the WIMAs?

4. I changed the polycarbonates at C15 and C16 to XICON 1429 series propylenes (exact value matches) for, I think, obvious reasons. Everyone agree?











Anyone wish to opine?