Hi Uncle,unclejed613 said:it would probably be a good idea to invest in or borrow an ESR tester for caps, there's a 100khz model out there that works well. or if you have an o-scope and a function generator that goes up to 100khz, you could build an ESR tester.
Can you give a link or build details?
I guess I have to throw two cents in here...
I have never noticed a difference in caps that I used in a power supply. I have replaced a number of them thru the years. I have used a number of brands and the result was always positive.
I have however noticed a difference and this would be a slight difference in smaller caps used in preamps and line stages. I have also learned over the years to check the caps I purchased to make sure that haven't been mismarked as to the value.
When I want to make sure the sonic signature will be the same I will replace with the same type and brand as the original. There is a good chance that the sound you have is quite good and cleaner than it was before.
A number of years ago I was introduced to using bypass caps in the power supply of power amplifiers. At first it made me think that I had lost some of the punch in the low end that I used to have. It took the logic and experience of Mr. DJK to make me see that I hadn't lost any low end punch but had awakened some of the never heard and missing midrange detail.
I would suggest that you take your equipment to a Hi Fi shop and have them run some frequency response tests and plot the response of these tests and come back and post the results.
I have never noticed a difference in caps that I used in a power supply. I have replaced a number of them thru the years. I have used a number of brands and the result was always positive.
I have however noticed a difference and this would be a slight difference in smaller caps used in preamps and line stages. I have also learned over the years to check the caps I purchased to make sure that haven't been mismarked as to the value.
When I want to make sure the sonic signature will be the same I will replace with the same type and brand as the original. There is a good chance that the sound you have is quite good and cleaner than it was before.
A number of years ago I was introduced to using bypass caps in the power supply of power amplifiers. At first it made me think that I had lost some of the punch in the low end that I used to have. It took the logic and experience of Mr. DJK to make me see that I hadn't lost any low end punch but had awakened some of the never heard and missing midrange detail.
I would suggest that you take your equipment to a Hi Fi shop and have them run some frequency response tests and plot the response of these tests and come back and post the results.
I changed ONLY the power supply filter caps in my 110 - and now it sounds like garbage.
Sorry! I didn't see this
Over the years, I noticed the sound seemed to harshen and the deepest bass gradually disappeared.
Memory if a funny animal ....there is a good chance your trying to remember something that was never there.
I have a friend who liked his Hafler pre very much but he was forced to scrap it
after several rounds of cleaning he had to realize that all connections, switches and attenuator was completely worn out
nothing last forever
after several rounds of cleaning he had to realize that all connections, switches and attenuator was completely worn out
nothing last forever

Solder choices are as important to the sonics as any parts choice. The original likely used 60/40 or 63/37. Try using the Cardas quad eutectic, and let the thing (amp and preamp) burn in for about 2 weeks.
Must... reuse... 20 yr old lytics - is this a joke?
Ok, the listening continues.
Plugged in my other Hafler 280 and immediately noticed the same thing. If I had to describe it, it's like the sound is running through an equalizer and the 20kHz band is set to -10. Weird and very, very irritating.
I'm hearing this mainly in the cymbals, chimes, and steel strings. All of the tinny 'metal' is gone from the sound. It sounds utterly lifeless and flat.
After 'The Flaming Lips' I tried Bob Dylan's 'Time Out Of Mind'. The sound is clear, don't get me wrong, but the edge is gone from Bob's voice and, as usual, the cymbals sound like they are coming from under a pillow. 'It's not dark yet, but it's gettin' there' - you said it Bob.
After that, I put on a recording that I know to be too bright, The Smith's 'Louder Than Bombs'. Even this sounded muted and was astonishingly lifeless considering the original recording was lacking to begin with.
At this point, the IC caps are on their way back in, and, yes, it rubs me the wrong way to put 20 year old lytics in anything - but I cannot live with this any longer than I have to! I can let the new stuff 'burn in' until next week when I will have time to put the old stuff back.
One last thing, I can hear this difference in the preamp with 2 different pairs of headphones, not just my amp/speaks.
Replies:
- To tinitus -- My 110 is in amazing physical condition. I have treated it gently and have never stored it. The only issue I have had was my dirty Tape Monitor selector which sounds fine now that I cleaned it. The volume and balance are dead quiet.
By the way, PROPS to soundvalves, you should see my 110 with its new knobs, it is so beautiful - like new.
- To KBK -- I use a good 60/40. I flux each joint thoroughly and clean with alcohol before making a fresh connection. I use a good soldering iron with a ceramic heater and if my finished joint isn't smooth and mirror finish, I do it again. Personally, I very much doubt that you can 'hear' solder...
Ok, the listening continues.
Plugged in my other Hafler 280 and immediately noticed the same thing. If I had to describe it, it's like the sound is running through an equalizer and the 20kHz band is set to -10. Weird and very, very irritating.
I'm hearing this mainly in the cymbals, chimes, and steel strings. All of the tinny 'metal' is gone from the sound. It sounds utterly lifeless and flat.
After 'The Flaming Lips' I tried Bob Dylan's 'Time Out Of Mind'. The sound is clear, don't get me wrong, but the edge is gone from Bob's voice and, as usual, the cymbals sound like they are coming from under a pillow. 'It's not dark yet, but it's gettin' there' - you said it Bob.
After that, I put on a recording that I know to be too bright, The Smith's 'Louder Than Bombs'. Even this sounded muted and was astonishingly lifeless considering the original recording was lacking to begin with.
At this point, the IC caps are on their way back in, and, yes, it rubs me the wrong way to put 20 year old lytics in anything - but I cannot live with this any longer than I have to! I can let the new stuff 'burn in' until next week when I will have time to put the old stuff back.
One last thing, I can hear this difference in the preamp with 2 different pairs of headphones, not just my amp/speaks.
Replies:
- To tinitus -- My 110 is in amazing physical condition. I have treated it gently and have never stored it. The only issue I have had was my dirty Tape Monitor selector which sounds fine now that I cleaned it. The volume and balance are dead quiet.
By the way, PROPS to soundvalves, you should see my 110 with its new knobs, it is so beautiful - like new.
- To KBK -- I use a good 60/40. I flux each joint thoroughly and clean with alcohol before making a fresh connection. I use a good soldering iron with a ceramic heater and if my finished joint isn't smooth and mirror finish, I do it again. Personally, I very much doubt that you can 'hear' solder...
IMO, this is beginning to sound like something else is wrong. Even if you can hear power supply caps, which I can't, it's beyond belief they can cause the radical difference you're describing. Find somebody with a test bench and at least verify the response of the unit.
I just looked up the schematic for the DH110, and noticed that is uses 3-terminal regulators. You need to be careful with these, as low ESR output caps can cause them to oscillate. Normal-style electrolytics like the Panasonic NHG series have enough ESR so that thay should be safe.
I agree with you, Conrad Hoffman, the only thing that keeps me listening is the fact that I pretested each cap and installed very carefully.
Thank you wrenchone, the unit is powered down now and too boring to listen to anyway.
Yes, it does use a couple voltage regulators, and the caps in conjunction with said regs are now very low impedence Panasonic FM caps, that is, the 50V 1000uF and 22uF. I did leave the unit on all night last night, and this morning it was completely cool to the touch with the exception of a bit of normal warmth over the transformer.
One of my inspirations for acquiring the ELNA SILMIC II caps for the rails is the fact that the schematic reads exactly this:
470mfd, low ESR electrolytic, 25V
Now that's from David Hafler Co. themselves.
I do find wrenchone's comment very helpful because I was thinking of trading out the ELNAs first, then the FMs if things did not improve.
At this point I will probably not power it up again and will switch the 4 power caps first. Thanks again to all who have made suggestions, I will keep you updated on this.
Thank you wrenchone, the unit is powered down now and too boring to listen to anyway.
Yes, it does use a couple voltage regulators, and the caps in conjunction with said regs are now very low impedence Panasonic FM caps, that is, the 50V 1000uF and 22uF. I did leave the unit on all night last night, and this morning it was completely cool to the touch with the exception of a bit of normal warmth over the transformer.
One of my inspirations for acquiring the ELNA SILMIC II caps for the rails is the fact that the schematic reads exactly this:
470mfd, low ESR electrolytic, 25V
Now that's from David Hafler Co. themselves.
I do find wrenchone's comment very helpful because I was thinking of trading out the ELNAs first, then the FMs if things did not improve.
At this point I will probably not power it up again and will switch the 4 power caps first. Thanks again to all who have made suggestions, I will keep you updated on this.
As I pointed out in a earlier post... take your equipment to a Hi Fi
shop and have them run a couple of tests and this will confirm if there is actually anything wrong. Have them graph the response and post it back here.
shop and have them run a couple of tests and this will confirm if there is actually anything wrong. Have them graph the response and post it back here.
AndrewT said:
Hi Uncle,
Can you give a link or build details?
Andrew,
I have the "Capacitor Wizard" made by Independence Electronics (USA). Sadly, they seem to have gone out of business. Another similar tool is at http://www.abra-electronics.com/products/catalog/CapAnalyzer-88A-Series-II-p-5243.html These are very valuable for quickly checking for open caps in a misbehaving circuit but may be of dubious value in audio comparisons. BTW, when I made my living at this the Panasonic NHG series was the lowest ESR reasonably priced general purpose cap that I had found.
Bob G.
I would suggest checking the bias. I've got a small tribe of Haflers around here and every one of them had gone lean before I got 'em. Do not underestimate the difference bias can make in tone, imaging, etc. Higher is always better, subject to limitations of power supply, heat sink, and so forth...but at the bare minimum, verify that the amp is at least up to factory spec. as they will drift over the years.
Grey
Grey
Halfer dude.
Some thoughts.
While some have justifiably focussed on the fact that there may be other reasons for the difference in hearing you perceive, I am of the opinion that the drastic difference you heard right after the change means that the caps "were" probably the reason. As far as the "mental" aspect goes, a typical DIY'e will be biased in favor of a given mod because he wants to succeed... and despite that, you did not like it means that it probably did actually sound worse.
Here's a theory... Low ESR caps provide the OP section with enough current that some sort of beta droop is occuring in the OP or driver stage as they strive to drive the load..... with softer caps, the rails sag more reducingdelivered power... Granted MOSFETS are relatively immune, still. A softer supply will sometimes sound better as the OP as the active devices are not taxed as much and perhaps may work in their linear range.... Just a thought...
In support of the theory... if the kick drums sound punchier but the highs and mids harsher, then it is a stronger possibility of this phenomenon IMHO.
On a side note: Do not underestimate the difference in sound quality you can get by upgrading the lytics in the crossovers of your speakers... merely increasing the voltage rating alone will do wonders... I had some old KEF's C95's? They had 25v caps in the crossover.. !!!! Needless to say they did not sound constricted once I went with 250v metallized polyprops (brand not important in this case, they were $3 each from Apex Jr.).
Some thoughts.
While some have justifiably focussed on the fact that there may be other reasons for the difference in hearing you perceive, I am of the opinion that the drastic difference you heard right after the change means that the caps "were" probably the reason. As far as the "mental" aspect goes, a typical DIY'e will be biased in favor of a given mod because he wants to succeed... and despite that, you did not like it means that it probably did actually sound worse.
Here's a theory... Low ESR caps provide the OP section with enough current that some sort of beta droop is occuring in the OP or driver stage as they strive to drive the load..... with softer caps, the rails sag more reducingdelivered power... Granted MOSFETS are relatively immune, still. A softer supply will sometimes sound better as the OP as the active devices are not taxed as much and perhaps may work in their linear range.... Just a thought...
In support of the theory... if the kick drums sound punchier but the highs and mids harsher, then it is a stronger possibility of this phenomenon IMHO.
On a side note: Do not underestimate the difference in sound quality you can get by upgrading the lytics in the crossovers of your speakers... merely increasing the voltage rating alone will do wonders... I had some old KEF's C95's? They had 25v caps in the crossover.. !!!! Needless to say they did not sound constricted once I went with 250v metallized polyprops (brand not important in this case, they were $3 each from Apex Jr.).
Hi HaflerFreak,
Sorry to hear you are really having a rough time with this. This is supposed to be like a rebirth, not purgatory.
Electrolytics are supposed to almost take a lifetime to burn in but something else has to be wrong or the amp would have sounded like garbage when you first got it for quite a while. Usually the biggest change in sound happens in the first 8 to 24 hours of burn in and in small increments up to 350 hours for the remaining percentage of change in sound.
Just in case you prefer to burn in your Caps from now on as you change them and decide on several different makes (but to make that comparison fairly, they should have many and an equal amount of hours on them), here is something easy and cheap to make. Panomaniac supplied this "Burn in Circuit" on another thread. Here is the link: http://www.diyaudio.com/forums/showthread.php?postid=1321014#post1321014
Good luck in solving this mystery, Check everything from the transformer to the rectifiers to grounds all the way to the binding posts! I have a DH500 since the early 80s and would like to see you, another Hafler owner, successful and more than happy again!
Regards//Keith
Sorry to hear you are really having a rough time with this. This is supposed to be like a rebirth, not purgatory.
Electrolytics are supposed to almost take a lifetime to burn in but something else has to be wrong or the amp would have sounded like garbage when you first got it for quite a while. Usually the biggest change in sound happens in the first 8 to 24 hours of burn in and in small increments up to 350 hours for the remaining percentage of change in sound.
Just in case you prefer to burn in your Caps from now on as you change them and decide on several different makes (but to make that comparison fairly, they should have many and an equal amount of hours on them), here is something easy and cheap to make. Panomaniac supplied this "Burn in Circuit" on another thread. Here is the link: http://www.diyaudio.com/forums/showthread.php?postid=1321014#post1321014
Good luck in solving this mystery, Check everything from the transformer to the rectifiers to grounds all the way to the binding posts! I have a DH500 since the early 80s and would like to see you, another Hafler owner, successful and more than happy again!
Regards//Keith
Happiness is a warm -- Hafler
I believe in truth. I believe in facts because knowledge is built upon other knowledge.
How can I deceive my favorite forum and those that seek knowledge in these pages? I cannot.
The truth is, when I performed my 'mod', I did it over the course of a week or so, Elnas first, then the FMs for the PS. The problem is, I also changed a signal cap that I entirely forgot to mention.
In the excitement of 'destroying' my equipment, I had forgotten C21. C21 is a .01uF PP bypass cap that bypasses one of those magic IC bipolars at the output of the line amp. The schematic calls for a 50V .01 PP, the Haffy came with a 100V axial standing on end. I switched it to a 63V WIMA FKP, supposedly for 'audio' use as they claim. Whatever. Cymbal killer.
Just for good measure I changed those 1000uF jobs back to the original IC caps and put the original PPs back at C21/121.
No other changes were made, that leaves the new Elnas and the FM caps as the 50V 22uF PS caps.
Sounds pretty much like it did before.
It doesn't sound like much at all 😉
As some of you have mentioned, a working PS is a working PS with pretty much any working caps, so this probably means DO NOT use WIMA FKP caps (hee hee).
Things seem to be back to normal, and I have achieved my goal regardless of whether I can hear the Elnas or not, I have extended the life of my equipment and that means a whole heckuvalot to me these days.
I'll try to attach a pic...
I believe in truth. I believe in facts because knowledge is built upon other knowledge.
How can I deceive my favorite forum and those that seek knowledge in these pages? I cannot.
The truth is, when I performed my 'mod', I did it over the course of a week or so, Elnas first, then the FMs for the PS. The problem is, I also changed a signal cap that I entirely forgot to mention.
In the excitement of 'destroying' my equipment, I had forgotten C21. C21 is a .01uF PP bypass cap that bypasses one of those magic IC bipolars at the output of the line amp. The schematic calls for a 50V .01 PP, the Haffy came with a 100V axial standing on end. I switched it to a 63V WIMA FKP, supposedly for 'audio' use as they claim. Whatever. Cymbal killer.
Just for good measure I changed those 1000uF jobs back to the original IC caps and put the original PPs back at C21/121.
No other changes were made, that leaves the new Elnas and the FM caps as the 50V 22uF PS caps.
Sounds pretty much like it did before.
It doesn't sound like much at all 😉
As some of you have mentioned, a working PS is a working PS with pretty much any working caps, so this probably means DO NOT use WIMA FKP caps (hee hee).
Things seem to be back to normal, and I have achieved my goal regardless of whether I can hear the Elnas or not, I have extended the life of my equipment and that means a whole heckuvalot to me these days.
I'll try to attach a pic...
Attachments
K-amps said:.. merely increasing the voltage rating alone will do wonders... I had some old KEF's C95's? They had 25v caps in the crossover.. !!!! Needless to say they did not sound constricted once I went with 250v metallized polyprops (brand not important in this case, they were $3 each from Apex Jr.).
Why would you think the voltage rating made the difference? Why not the colour? You have replaced electrolytics with polyprops and you expect the voltage rating was responsible for the improvement?!
It sounds more likely that the little FKP cap did not match well with the rest of the system, than the PSU messing up frequency response... Leave the PSU as is ans swap out that wima cap with others... I can for instance recommend trying roederstein caps
HaflerFreak
After looking at your chassis picture I might suggest that you also use cable ties on the caps to draw them closer together. This will have the effect of extending the low frequency response. White ties seem to make for better sound.
After looking at your chassis picture I might suggest that you also use cable ties on the caps to draw them closer together. This will have the effect of extending the low frequency response. White ties seem to make for better sound.
“It has long been an axiom of mine that the little things are infinitely the most important” A. C. Doyle Sr.
Re: Happiness is a warm -- Hafler
I have a DH-110, gotta say that I was not impressed with the flimsy chassis and mechanical work, hate all the tin plated connectors. The connectors got so bad on mine that one channel would drop out. If yours are only mildly corroded they might be a source of distortion.
The timing caps in the mute circuit got so bad that it would not unmute. I replaced them and agree that all the electrolytics should be replaced. I'd be surprised if the issue with yours was the caps, on the other hand there are some large electrolytics in the signal path. I'd hit the connectors with some DeOxit before blaming the caps.
You should do the null test on your XL-280, this removes the subjective considerations for that unit.
I would also suggest that you test with some old good reliable source material that you are familiar with, because much of the available source material is certainly the weak link in the chain.
Pete B.
HaflerFreak said:I believe in truth. I believe in facts because knowledge is built upon other knowledge.
How can I deceive my favorite forum and those that seek knowledge in these pages? I cannot.
The truth is, when I performed my 'mod', I did it over the course of a week or so, Elnas first, then the FMs for the PS. The problem is, I also changed a signal cap that I entirely forgot to mention.
In the excitement of 'destroying' my equipment, I had forgotten C21. C21 is a .01uF PP bypass cap that bypasses one of those magic IC bipolars at the output of the line amp. The schematic calls for a 50V .01 PP, the Haffy came with a 100V axial standing on end. I switched it to a 63V WIMA FKP, supposedly for 'audio' use as they claim. Whatever. Cymbal killer.
Just for good measure I changed those 1000uF jobs back to the original IC caps and put the original PPs back at C21/121.
No other changes were made, that leaves the new Elnas and the FM caps as the 50V 22uF PS caps.
Sounds pretty much like it did before.
It doesn't sound like much at all 😉
As some of you have mentioned, a working PS is a working PS with pretty much any working caps, so this probably means DO NOT use WIMA FKP caps (hee hee).
Things seem to be back to normal, and I have achieved my goal regardless of whether I can hear the Elnas or not, I have extended the life of my equipment and that means a whole heckuvalot to me these days.
I'll try to attach a pic...
I have a DH-110, gotta say that I was not impressed with the flimsy chassis and mechanical work, hate all the tin plated connectors. The connectors got so bad on mine that one channel would drop out. If yours are only mildly corroded they might be a source of distortion.
The timing caps in the mute circuit got so bad that it would not unmute. I replaced them and agree that all the electrolytics should be replaced. I'd be surprised if the issue with yours was the caps, on the other hand there are some large electrolytics in the signal path. I'd hit the connectors with some DeOxit before blaming the caps.
You should do the null test on your XL-280, this removes the subjective considerations for that unit.
I would also suggest that you test with some old good reliable source material that you are familiar with, because much of the available source material is certainly the weak link in the chain.
Pete B.
- Status
- Not open for further replies.
- Home
- Amplifiers
- Solid State
- How to destroy your high-end audio equipment...