My preamp has high DC offset voltage (9.7 and 10.4 VDC, left and right ) at the output terminals. It is a Hafler DH110. I have not found anything on whether this is a DC coupled preamp. The schematic shows what looks like a coupling film capacitor at C19 (470 nfd, 250V) , but it seems unlikley that two film capacitors would fail. I pulled one and it was not shorted and metered at 490 nfd. I checked/replaced all three electrolytics in the line amp section for one channel, and this did not help.
I noticed that the sound was not as clear as it normally is. Another problem is a big transient at turn-on and turn-off that can blow speaker fuses. These were the motivations for checking.
The muting circuit seems to be working in that it takes about 5 sec before voltage comes up to about 10 at the outputs. However, just before it does, there is a sharp spike to about 20 volts. Could a faulty muting circuit cause high offset?
The power amp has a 2 mfd coupling cap at the input. Would this filter all of the DC from the preamp? Even so, I would still like to reduce the DC offset as much as possible.
Perhaps another possibility is unbalanced transistors, but it seems unlikley that both channels would have this problem.
A friend put it on his scope, and noticed a 10 V peak to peak ripple current at the output for both channels. Probes into the power supply indicated that it was okay (low ripple, voltage at specification). The problem seems to be in the line amp section, but an almost equal voltage for each channel suggests that it happens before the line amp. Volume and balance controls have no effect.
Any suggestions?
I noticed that the sound was not as clear as it normally is. Another problem is a big transient at turn-on and turn-off that can blow speaker fuses. These were the motivations for checking.
The muting circuit seems to be working in that it takes about 5 sec before voltage comes up to about 10 at the outputs. However, just before it does, there is a sharp spike to about 20 volts. Could a faulty muting circuit cause high offset?
The power amp has a 2 mfd coupling cap at the input. Would this filter all of the DC from the preamp? Even so, I would still like to reduce the DC offset as much as possible.
Perhaps another possibility is unbalanced transistors, but it seems unlikley that both channels would have this problem.
A friend put it on his scope, and noticed a 10 V peak to peak ripple current at the output for both channels. Probes into the power supply indicated that it was okay (low ripple, voltage at specification). The problem seems to be in the line amp section, but an almost equal voltage for each channel suggests that it happens before the line amp. Volume and balance controls have no effect.
Any suggestions?
Do you have the schematic ???
My first thought was that one supply rail was down, but since you have meassured the supply with good results, my next idea is that something is wrong at the input circuit... Malfunctioning opamp, fet or trannie....
My first thought was that one supply rail was down, but since you have meassured the supply with good results, my next idea is that something is wrong at the input circuit... Malfunctioning opamp, fet or trannie....
Thanks for the reply. I have the original manual with the schematic, and Hafler provides pdf's of the manual that includes the schematic. Should I send you a copy? The file is 950K.
Hi mmerig,
Your preamp has capacitively coupled outputs. The muting section (Q15) could cause your DC offset. The easy way to check this is to remove the transistors (FET's) and turn the unit back on. Measure the output voltage.
If that's not the cause, check the voltage on each side of the output caps. Also check that your ground reference is sitting at ground potential.
-Chris
Your preamp has capacitively coupled outputs. The muting section (Q15) could cause your DC offset. The easy way to check this is to remove the transistors (FET's) and turn the unit back on. Measure the output voltage.
If that's not the cause, check the voltage on each side of the output caps. Also check that your ground reference is sitting at ground potential.
-Chris
Anatech, your advice on the muting circuit helped, thank you. I pulled Q15 from the left channel, and DC offset is now 0.00 mV on the left. It is slightly higher (11.2 volts versus 10.4) on the right channel.
My ohmeter check of Q15 shows that it is bad. I'm going to order and replace both Q15 and Q115, (both are J112 FET), but will order two extra just in case the IC3 is faulty and had broken both j112's.
My original hunch was that IC3 was bad, and I ordered one a week ago. But when I saw the delayed voltage rise at the output after turn-on, I figured the IC3 was okay. But after seeing no delay in the voltage rise at the power supply, it increased my suspicion about the muting circuit. Pulling Q15 helped confirm things.
Luckily, these parts are only about 50 cents apiece.
My ohmeter check of Q15 shows that it is bad. I'm going to order and replace both Q15 and Q115, (both are J112 FET), but will order two extra just in case the IC3 is faulty and had broken both j112's.
My original hunch was that IC3 was bad, and I ordered one a week ago. But when I saw the delayed voltage rise at the output after turn-on, I figured the IC3 was okay. But after seeing no delay in the voltage rise at the power supply, it increased my suspicion about the muting circuit. Pulling Q15 helped confirm things.
Luckily, these parts are only about 50 cents apiece.
Replacing Q15 and Q115 fixed the problem. I did not have to replace IC3. It has worked fine for a few days so far. Thanks for the suggestion.
I can only guess why both of these transistors failed, and I think it may have been static electricity. I live in a cold, dry climate, heat with wood, and usually wear wool. Unless I ground myself on the wood stove before I touch anythingg else metal, I usually get a small shock.
I can only guess why both of these transistors failed, and I think it may have been static electricity. I live in a cold, dry climate, heat with wood, and usually wear wool. Unless I ground myself on the wood stove before I touch anythingg else metal, I usually get a small shock.
Hi mmerig,
Those FET's aren't very happy with the high reverse bias voltage. You should check it as it may need clamping. Better yet. Yank out those parts and install a relay to short the signals to ground during mute.
Failure of these parts is reasonably common in many brands. Just think, there are affecting the sound before they actually "blow".
Hi Giaime,
Yup, it has a phone stage.
Those FET's aren't very happy with the high reverse bias voltage. You should check it as it may need clamping. Better yet. Yank out those parts and install a relay to short the signals to ground during mute.
Failure of these parts is reasonably common in many brands. Just think, there are affecting the sound before they actually "blow".
Hi Giaime,
Yup, it has a phone stage.
Hi Giame:
As anatech mentioned, the hafler dh110 preamp has a phono stage. It will not handle low-output moving coil cartridges, but hafler made a few head amplifiers that could be wired in for this. There are instructions on tuning the capacitance to match moving magnic cartridges.
I like it the sound; it seems neutral and it is very quiet (low noise).
Hi anatech:
Hmm, I thought it was back to normal. Where should I measure the bias voltage, at Q15? Can I balance it out somehow with a trimpot somewhere? Could IC3 be part of the problem?
I'll do some research on clamping. I am new at amp repair.
Thanks again.
As anatech mentioned, the hafler dh110 preamp has a phono stage. It will not handle low-output moving coil cartridges, but hafler made a few head amplifiers that could be wired in for this. There are instructions on tuning the capacitance to match moving magnic cartridges.
I like it the sound; it seems neutral and it is very quiet (low noise).
Hi anatech:
Hmm, I thought it was back to normal. Where should I measure the bias voltage, at Q15? Can I balance it out somehow with a trimpot somewhere? Could IC3 be part of the problem?
I'll do some research on clamping. I am new at amp repair.
Thanks again.
Hi mmerig,
The gates of the fets are brought to a high reverse voltage when unmuted. This voltage may be on the verge of the gate source breakdown voltage. There is no adjustment for this. The trick is to bias the fet off for the highest signal peaks. That's possibly a tall order.
A relay doesn't suffer from this and almost everyone who has pulled the fets for a relay reports better sound quality. At least relays won't distort the sound at all.
-Chris
The gates of the fets are brought to a high reverse voltage when unmuted. This voltage may be on the verge of the gate source breakdown voltage. There is no adjustment for this. The trick is to bias the fet off for the highest signal peaks. That's possibly a tall order.
A relay doesn't suffer from this and almost everyone who has pulled the fets for a relay reports better sound quality. At least relays won't distort the sound at all.
-Chris
Hafler DH110 preamplifier muting FETs
Hi everybody,
I resurrected this old thread because I have a question concerning the muting circuit of the DH110.
I have been listening to my 110 a lot lately having finally found the perfect PS caps (rails: Elna Tonerex + PS:Nichicon VZ) - nothing else modified, it sounds incredible
Only problem: Occasionally, it seems like the right channel becomes a little 'quieter?' than the left channel. There is no distortion of any kind and it happens randomly. It has been A/B tested back to the preamp. It seems to be most noticeable with a laugh track on TV DVDs.
I am pretty sure it might be the muting JFETs.
The original schematic calls for 2SK147 or J112 FETs -
My, factory assembled 110, has NEC K163's.
- I have on hand some 2SK170BL's from AmpsLab or...
- I can order Fairchild J112's for .15 each or...
- I could order 20 K163's from AmpsLab for $20 (original replacement)
Would all/any of these FETs be appropriate?
😕 Is there any advantage to a higher quality FET (K170/K147) compared to a lower quality (J112) in a muting circuit?
Any reason to stick strictly with the K163's?
Thank you to anyone who actually reads/deciphers all these numbers,
I really appreciate any suggestions.
Also, I would love to hear more about this:
Hi everybody,
I resurrected this old thread because I have a question concerning the muting circuit of the DH110.
I have been listening to my 110 a lot lately having finally found the perfect PS caps (rails: Elna Tonerex + PS:Nichicon VZ) - nothing else modified, it sounds incredible
Only problem: Occasionally, it seems like the right channel becomes a little 'quieter?' than the left channel. There is no distortion of any kind and it happens randomly. It has been A/B tested back to the preamp. It seems to be most noticeable with a laugh track on TV DVDs.
I am pretty sure it might be the muting JFETs.
The original schematic calls for 2SK147 or J112 FETs -
My, factory assembled 110, has NEC K163's.
- I have on hand some 2SK170BL's from AmpsLab or...
- I can order Fairchild J112's for .15 each or...
- I could order 20 K163's from AmpsLab for $20 (original replacement)
Would all/any of these FETs be appropriate?
😕 Is there any advantage to a higher quality FET (K170/K147) compared to a lower quality (J112) in a muting circuit?
Any reason to stick strictly with the K163's?
Thank you to anyone who actually reads/deciphers all these numbers,
I really appreciate any suggestions.
Also, I would love to hear more about this:
listen to the anatech ...
believe the anatech ...
but seriously - use a relay instead of any jfet - you won't be sorry.
mlloyd1
believe the anatech ...
but seriously - use a relay instead of any jfet - you won't be sorry.
mlloyd1
Mlloyd!
It is an honor sir, I always look forward to your input.
At the risk of sounding a bit stupid, I understand what relays do and how one would be connected here, but are you referring to a 'drop-in' type of unit where the JFET is mounted or is there a bit of wiring? I have never shopped relays in my big Mouser catalog... I'll get to that.
Are any other changes needed? Gate resistor, etc.
I'm afraid I would need specific guidance as to any re-engineering of the circuit.
As far as a relay goes, are there any crucial specs I will need to pay attention to?
I assume it has to safely pass 35 to 50 volts. What about the switch turn on/off voltage?
Are there relays that will physically drop-in or closely fit this app?
I know, so many questions!!
I feel guilty asking, I get such priceless info 🙂
It is an honor sir, I always look forward to your input.
At the risk of sounding a bit stupid, I understand what relays do and how one would be connected here, but are you referring to a 'drop-in' type of unit where the JFET is mounted or is there a bit of wiring? I have never shopped relays in my big Mouser catalog... I'll get to that.
Are any other changes needed? Gate resistor, etc.
I'm afraid I would need specific guidance as to any re-engineering of the circuit.
As far as a relay goes, are there any crucial specs I will need to pay attention to?
I assume it has to safely pass 35 to 50 volts. What about the switch turn on/off voltage?
Are there relays that will physically drop-in or closely fit this app?
I know, so many questions!!
I feel guilty asking, I get such priceless info 🙂
Relay stuff
Ok, at the risk of talking to myself here,
I went relay shopping and at least I can ask the right questions.....
Should I be looking for a SPST relay for each channel?
Here's my best guess so far:
At turn on, the comparator circuit in the mute section energizes the JFET gate in order to short the signal + to ground through the DS. As voltages stabilize, the gate is turned off and the signal bypasses the open FET DS path.
To wire in a relay, I wire the coil to the JFET gate lead and ground, wire the signal across one end of the SPST switch, and the other side of the switch to ground.
My question now is:
Do I have anything backwards?
Would the coil voltage be 5V, 12V? I think.. 5?
I assume it is PCB low signal relay,
how do these look?
http://www.mouser.com/catalog/634/1654.pdf
I would really like to do this mod if I can do it right.
Ok, at the risk of talking to myself here,
I went relay shopping and at least I can ask the right questions.....
Should I be looking for a SPST relay for each channel?
Here's my best guess so far:
At turn on, the comparator circuit in the mute section energizes the JFET gate in order to short the signal + to ground through the DS. As voltages stabilize, the gate is turned off and the signal bypasses the open FET DS path.
To wire in a relay, I wire the coil to the JFET gate lead and ground, wire the signal across one end of the SPST switch, and the other side of the switch to ground.
My question now is:
Do I have anything backwards?
Would the coil voltage be 5V, 12V? I think.. 5?
I assume it is PCB low signal relay,
how do these look?
http://www.mouser.com/catalog/634/1654.pdf
I would really like to do this mod if I can do it right.
Although Chris' (anatech's) advice is greatly appreciated, I did not replace the jfets with relays.
As I understand it, the jfet portion of the muting circuit in the Hafler DH110 is wired in parallel with the outputs, and when activated (muted), the signal is grounded. Once the PS voltages stabilize, the jfet switch is turned off (open), and the output signal goes through the normal path (a wire). The D-S path of the jfet is still in parallel with the signal to ground when unmuted, but I figured this would have a negligible effect on sound.
Some fet muting circuits have the fet in series with the signal, and I assume they are switched on when unmuted. This circuit would affect the sound much more, as the fet is a big part of the unmuted signal path; perhaps this is what Chris was referring to.
The J112's G-S or G-D maximum voltage is -35 V. The Hafler's PS rails are 23 V each, so the jfet is probably not too stressed. One of mine failed (the original reason for the thread), but I have not had a problem since. I soldered in transistor sockets for the jfets in case I have to replace them again ( I have a DH102 with its own PS wired in the preamp, and this makes access to the bottom of the main PCB difficult).
As I understand it, the jfet portion of the muting circuit in the Hafler DH110 is wired in parallel with the outputs, and when activated (muted), the signal is grounded. Once the PS voltages stabilize, the jfet switch is turned off (open), and the output signal goes through the normal path (a wire). The D-S path of the jfet is still in parallel with the signal to ground when unmuted, but I figured this would have a negligible effect on sound.
Some fet muting circuits have the fet in series with the signal, and I assume they are switched on when unmuted. This circuit would affect the sound much more, as the fet is a big part of the unmuted signal path; perhaps this is what Chris was referring to.
The J112's G-S or G-D maximum voltage is -35 V. The Hafler's PS rails are 23 V each, so the jfet is probably not too stressed. One of mine failed (the original reason for the thread), but I have not had a problem since. I soldered in transistor sockets for the jfets in case I have to replace them again ( I have a DH102 with its own PS wired in the preamp, and this makes access to the bottom of the main PCB difficult).
Hello and thank you to mmerig for responding,
Since my last post I have been doing my homework, reading my library books, because I am well aware that I am far more ignorant than dumb 😉
Excuse my impulsive questions, I sometimes forget that it is DIY audio... So much for the better my friends, so much for the better.
I researched my relays. I found a nice Omron SPST-NC that I thought could work, but after all was said and done do you know what happened? I sat there and studied the schematic - over and over, a work of art to this mere mortal - and I discerned much of the premise of what you stated mmerig.
The Jfets, if functioning, would (should?) have little effect on signal integrity in this case.
I listened to my DH-110 after my last post, and it hasn't 'acted up' since. To tell the truth, I suspect my cheap Sony DVD player and it's compression of the Dolby Surround track may be the source of my problems. I have very sensitive hearing and notice the slightest changes in spatial accuracy.
My 18 year old, totally untouched (unserviced/modified), Harman Kardon HD7450 CD player makes some very beautiful sounds through my DH110/XL280 Mg2.6/R combo.
P.S.
Have to mention one more time.
I recapped the power section of my DH110. I tried many, many cap combos, seriously, you don't want to spend the money that I did. After trying a LOT of caps in the rail and main PS sections I found these wonderful beauties:
Hafler DH-110 rail caps:
(8x matched) ELNA 35V 470uF TONEREX from Mouser.
Hafler DH-110 Main PS caps:
(2x matched) Nichcion VZ 50V 1000uF
(2x matched) Nichicon VZ 50V 22uF
muting circuit:
Nichicon 35V 4.7uF BP
Don't touch any other cap in that pre (I would not touch that phono section unless I had to) and that sucker will sound beautiful.
Also, replaced the original carbon comp .1A type 313 fuse with a new carbon film version.
Sounds so, so nice.
Since my last post I have been doing my homework, reading my library books, because I am well aware that I am far more ignorant than dumb 😉
Excuse my impulsive questions, I sometimes forget that it is DIY audio... So much for the better my friends, so much for the better.
I researched my relays. I found a nice Omron SPST-NC that I thought could work, but after all was said and done do you know what happened? I sat there and studied the schematic - over and over, a work of art to this mere mortal - and I discerned much of the premise of what you stated mmerig.
The Jfets, if functioning, would (should?) have little effect on signal integrity in this case.
I listened to my DH-110 after my last post, and it hasn't 'acted up' since. To tell the truth, I suspect my cheap Sony DVD player and it's compression of the Dolby Surround track may be the source of my problems. I have very sensitive hearing and notice the slightest changes in spatial accuracy.
My 18 year old, totally untouched (unserviced/modified), Harman Kardon HD7450 CD player makes some very beautiful sounds through my DH110/XL280 Mg2.6/R combo.
P.S.
Have to mention one more time.
I recapped the power section of my DH110. I tried many, many cap combos, seriously, you don't want to spend the money that I did. After trying a LOT of caps in the rail and main PS sections I found these wonderful beauties:
Hafler DH-110 rail caps:
(8x matched) ELNA 35V 470uF TONEREX from Mouser.
Hafler DH-110 Main PS caps:
(2x matched) Nichcion VZ 50V 1000uF
(2x matched) Nichicon VZ 50V 22uF
muting circuit:
Nichicon 35V 4.7uF BP
Don't touch any other cap in that pre (I would not touch that phono section unless I had to) and that sucker will sound beautiful.
Also, replaced the original carbon comp .1A type 313 fuse with a new carbon film version.
Sounds so, so nice.
- Status
- Not open for further replies.