Hafler XL280 mods gain/low frequency cutoff problem - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 22nd August 2007, 08:00 AM   #1
diyAudio Member
 
HaflerFreak's Avatar
 
Join Date: Jun 2007
Location: Westmont, IL
Default Hafler XL280 mods gain/low frequency cutoff problem

Greetings DIYers,
I think I have an interesting problem that would be fairly easy for some of you to figure out.
Recently, I got the DIY bug with my 19 year old Hafler XL-280 and decided to replace the electrolytics with the exception of the big 7800uF jobs. As usual, even after much reading, I rushed into the job and replaced the caps with Nichicon/XICON caps. Powered the unit up with no VARIAC etc. and - all was well. It actually sounded pretty much like I had always remembered. For the record, one side always ran a little hotter and occasionally this amp sounded a bit harsh to my ears.

Anyhow, I was listening to it one day and the right side just sort of fizzled. It did not go totally out, but volume was reduced and it was a bit distorted. I switched off and let the unit cool, and upon powering up again the right side was definitely lacking in bass. The output sounded clear and the mids and highs seemed all there, but no bass.

I came to the forum and found this:
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 is inseries 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.

Someone suggested that the 1000uF or 2.2k R might have blown, so...
I decided that if I was taking the thing apart again, I would do it right.
After quite a bit more reading, I purchased Panasonic FC caps, new PS diodes, and some other high quality caps for the driver board. I got a Variac and made a current limiting extension chord with a lightbulb socket. I splurged for a BK810C precision cap tester and cherry picked 2 47pF polystyrenes for C1 which had been micas (propylenes in the schematic), I was rollin'.

This was pretty much what I wound up with

For the right (bad) side, I tested the 1000uF, the 2.2K resistor and various other parts - found nothing wrong except the 47 Ohm carbon comps at R34,35 had nearly burst into flame judging by the melted areas underneath them. Nevertheless, they tested 47.1 and 56.3 Ohm - I replaced them with matched 47.1 Ohm 1/2W flameproof metal-films.
Tested the big main caps since I was there, they were all 8000+ uF - don't know about ESR or leakage.
Assemebled all my good stuff with a few experiments like XICON 1431 caps and an Orange Drop at C4.
Powered her up and she looked good, here she is. - this is the 'bad' side.
This is the good side

So everything looked good, and I hooked up the Maggies.
Right side sounds crystal clear but a little low on gain and definitely lacking the low bass.

MY QUESTION IS:
Given the schematic and parts list below, what is it! This is driving me nuts. I think all rare things (like FETS) are Ok given the clarity of sound and the good voltage and bias readings. What in the gain and low frequency cutoff circuit could do this. The ONE cap that I did not test was C13, lemme guess, that's probably it - damn Murphys Law.

BTW in my rebuild, I found a blown electrolytic bypass cap (C15) on the GOOD side, funny.
XL280 Schematic
XL280 Power Supply schematic
XL280 Driver Parts Layout/Voltages
XL280 Parts List

I would really love to hear any stories from people that have messed with a XL280, this amp has great sentimental value for me and I would love to hear it sing again.
  Reply With Quote
Old 22nd August 2007, 05:42 PM   #2
mlloyd1 is offline mlloyd1  United States
diyAudio Member
 
mlloyd1's Avatar
 
Join Date: Feb 2001
Location: illinois
haflerfreak:

sorry to hear about your troubles. i've always liked the XL-280, too.

the stupid questions first before the smart guys respond:

0. are you sure the speaker is still good? Maybe switch them to make sure. Don't laugh - I could tell you some funny stories ...

1. did you check the solder connections (especially where you replaced parts)? Do more than visually inspect; wiggle the parts while checking for electrical continuity, especially R26 and C8.

2. i am worried about the burned areas under R34 and R35. i assume you did not see this under the good side? Are C16-C19 OK? Any way you could provide some voltage measurements for the transistors on the bad side (and maybe compare to same on the good side)?

3. Are you able to measure +/- rail currents (lifting the rail fuse makes it easier - BTW, you checked that these fuses are good, right)? I wonder if the output mosfets are even active - the rail fuses don't disable the driver board when they open ...

all for now,
mlloyd1
  Reply With Quote
Old 22nd August 2007, 06:20 PM   #3
PSz. is offline PSz.  United States
diyAudio Member
 
PSz.'s Avatar
 
Join Date: Jul 2003
Location: California
output fuses?
  Reply With Quote
Old 22nd August 2007, 08:04 PM   #4
mlloyd1 is offline mlloyd1  United States
diyAudio Member
 
mlloyd1's Avatar
 
Join Date: Feb 2001
Location: illinois
Quote:
Originally posted by PSz.
output fuses?
if this was a DH200 (maybe DH-220 also), i would offer the same guess, too.
hafler used to put the speaker fuse in the feedback loop and put a resistor in parallel to maintain the dc path, so you could still hear some sound if that fuse opened. however, that's not done anymore according to the schematic for the XL-280.

mlloyd1
  Reply With Quote
Old 22nd August 2007, 09:30 PM   #5
PSz. is offline PSz.  United States
diyAudio Member
 
PSz.'s Avatar
 
Join Date: Jul 2003
Location: California
I had an output fuse in my xl280 go "softly" like this. It still played, just not well. I replaced both fuses and all was well.

best regards,
psz.
  Reply With Quote
Old 22nd August 2007, 09:42 PM   #6
diyAudio Member
 
HaflerFreak's Avatar
 
Join Date: Jun 2007
Location: Westmont, IL
Hey mlloyd1, thanks so much for the reply. You asked the right questions. When the right side fizzled, I tried switching inputs and sources to narrow it down. I would absolutely swear that I tried switching the speakers when this all happened.
I must say, I'm a bit embarrassed.
I just switched the old magnepans and I am seriously not sure that I can believe my ears.
I cranked that balance over to R and that left speak just sings.

I have 2 other stupid posts on this site, one that wound up being a blown DMM fuse and another about a jittery bias pot (that I just adjusted little by little with the power off each time), but this takes the cake!

I have to go to work right now (almost late), but I have an old pair of JBLs that I will pull out tonight to verify this and ENJOY my XL280!

Will post listening stories about the caps too.


Thanks again guys.
  Reply With Quote
Old 23rd August 2007, 08:26 AM   #7
diyAudio Member
 
HaflerFreak's Avatar
 
Join Date: Jun 2007
Location: Westmont, IL
Default No, it's the amp

Hey everyone,
Switched over to a pair of old JBLs using brand new wire, it's the amp. I was pretty sure I had verified that it wasn't the speakers before I took stuff apart.
well, at least the Magnepans are Ok and I wasn't THAT boneheaded.

But this leaves us with the original low frequency cutoff problem...

I can assure you that it is not a bad solder connect, nor is it fuses, and, of course, not the speakers.

One of the reasons for my apparent confusion in the diagnosis is that it is somewhat subtle. The soundstage is pure and clear and focused - but the bottom couple octaves of bass are missing from the right channel.
I have checked C8, R26 and most, if not all, of the resistors and diodes. Everything fine.

I am honestly a bit paranoid to take voltage readings off of transistor legs etc. while this thing is powered up, even though I have a voltage chart for the driver board. I am a cautious and ignorant newbie who just wants his amp to go but is too poor to get it analyzed by a shop (not to mention I don't trust anyone these days that claims to be educated).

All I can tell you guys is that R34 and R35 melted pits about half way through the board - my soldering iron doesn't even melt this board!!!!
  Reply With Quote
Old 23rd August 2007, 03:41 PM   #8
diyAudio Member
 
HaflerFreak's Avatar
 
Join Date: Jun 2007
Location: Westmont, IL
To mlloyd1:

Sorry about my sluggish responses, I work all the time.

The polycarbs at C15,C16 tested fine, I have replaced the electrolytics at C17,C18 twice and I have always suspected my earlier attempt using XICON 100uFs and no Variac/lightbulb may have had something to do with R34,35 getting hot. To be honest, I don't know if those resistors got hot with the original 19 year old IC caps or with the new ones I have tried.

You said:
3. Are you able to measure +/- rail currents (lifting the rail fuse makes it easier - BTW, you checked that these fuses are good, right)? I wonder if the output mosfets are even active - the rail fuses don't disable the driver board when they open ...

I thought the bias reading from the fuse clip was the rail current?!? It is a fairly steady 285mA.

BTW another observation -
I have noticed that when I power down with source at low volume and drain off the caps, the 'bad' side goes silent almost right away while the good side plays for 10 seconds or so.


Thanks for any feedback, I hate 'just replacing stuff'...
  Reply With Quote
Old 23rd August 2007, 09:08 PM   #9
mlloyd1 is offline mlloyd1  United States
diyAudio Member
 
mlloyd1's Avatar
 
Join Date: Feb 2001
Location: illinois
haflerfreak:

don't apologize; between my day job and honey-do lists, i have very little time for play these days, too.

sorry if i wasn't clear. looking at the schematic, measuring current through F1 or F2 yields currents through the output mosfets. somewhere around 300mA is ok (Hafler usually wants about 100mA per mosfet, so 285mA is in the ballpark). i assume you measured both sides?

however, this doesn't help tell us how much current the driver board is using, though. if R34 and R35 have burned areas, something on the driver board is (or used to) draw a lot more current than it should have.

hmmm, i forgot the other stupid question: you did check carefully to make sure the replaced electrolytic caps were installed with correct polarities, right? caps with large voltages across them usually inform you of their unhappiness at being connected incorrectly with a noisy celebration, complete with smelly confetti.

so, what voltage measurements can you do?

mlloyd1
  Reply With Quote
Old 23rd August 2007, 10:34 PM   #10
diyAudio Member
 
HaflerFreak's Avatar
 
Join Date: Jun 2007
Location: Westmont, IL
Oh yeah, I double and triple check electrolytics before I power them up. I have never put one in wrong, and I hope to avoid it in the future.

I actually did not measure the bias from both sides and the B+ 285mA is from my most recent adjustments. Before I played with it, I just assumed it was biased high (over 300mA) because it always ran hotter than the other side and it now idles decidedly cooler than ever before. It worked like that for years...

When I redid all the caps, I reset bias on both sides to 285mA with a Variac and now both heatsinks seem to warm up almost identically - I assume that is good. There does not seem to be anything inclined to meltdown at the moment, but I haven't been running the amp for very long periods, just enough for testing etc.

Well, I can see we are going to need some numbers to work with, so I will try to get the cover off and take voltage and current readings for various things that I feel safe accessing. Today is my 1 day off, so I might be able to report later tonight hopefully.

Oh yeah, mlloyd1,
I can't quite describe the feeling to you when I suspected my $1000 each Magnepan MG2.6/R might be the culprit and not my $600 XL280!!
It was actually kind of funny. Talk to you later
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Hafler XL280 Thermal/Heat Switch? RnR Solid State 3 16th July 2009 04:33 PM
Hafler XL280-went kaput echelonphoto Solid State 8 14th January 2009 12:18 AM
Frequency Cutoff habsrock93 Full Range 2 21st August 2007 01:32 PM
help repairing Hafler xl280 Truckee_Tango Solid State 25 26th August 2006 06:08 PM
hafler xl280 amp info vacuumhead Solid State 31 30th September 2005 01:45 AM


New To Site? Need Help?

All times are GMT. The time now is 11:41 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2