Hello All,
New member, not at the level of expertise of many of you. Please forgive my dumb questions.
I recently bought a completely original DH-200 ($90) with a slightly elevated hiss in one channel. I've replaced all the electrolytics and cleaned the output mosfet pins on both boards. The noise is diminished but still there. Any suggestions on my next step?
Also, I wondered if anyone could explain why the right input is isolated from the chassis and has a resisitor running to ground? What purpose does the resistor serve? Why can't the negative of the RCA be grounded o the chassis directly?
Thanks in advance for any info....
New member, not at the level of expertise of many of you. Please forgive my dumb questions.
I recently bought a completely original DH-200 ($90) with a slightly elevated hiss in one channel. I've replaced all the electrolytics and cleaned the output mosfet pins on both boards. The noise is diminished but still there. Any suggestions on my next step?
Also, I wondered if anyone could explain why the right input is isolated from the chassis and has a resisitor running to ground? What purpose does the resistor serve? Why can't the negative of the RCA be grounded o the chassis directly?
Thanks in advance for any info....
Hey neddyboy. Hiss one channel??? I think that would be the differential stage. That would be the first place to start.
Some other questions:
Is it just a white noise type of hiss, or does it burst into pops and crackles?
Is the volume the same left compared to right?
So the order and stuff I'd check is:
1. differential stage.
2. Class A (VAS) transistor(s) ...I think that thing is full complementary.
3. Feddback shunt capacitor (electrolytic)
I think I have the schematic. Do you have? Maybe I could help ID some of the components??
Some other questions:
Is it just a white noise type of hiss, or does it burst into pops and crackles?
Is the volume the same left compared to right?
So the order and stuff I'd check is:
1. differential stage.
2. Class A (VAS) transistor(s) ...I think that thing is full complementary.
3. Feddback shunt capacitor (electrolytic)
I think I have the schematic. Do you have? Maybe I could help ID some of the components??
Hi neddyboy,
mrshow4u is correct. It could be either differential pair or a Vas transistor.
Here's the rub. The two NPN and PNP transistors need to be matched. Then they need to be matched together. So that is 4 transistors with the same hFE. If you don't have a way of matching transistors from a group, you may want to have it serviced by a good tech. You will need to buy a bunch of transistors to select from. ECG and NTE parts are no flyers. If you attempt to use those, you have lost the right to own any audio equipment. TV's are fine.
It is possible to have a noisy resistor. I've seen that before too. Try to keep an open mind.
Now, does the noise get louder as it gets warmer, or the reverse? Does temperature matter? Do you have a digital voltmeter? If so, measure the DC offset across the speaker terminals with the amp warmed up, no signal.
-Chris
mrshow4u is correct. It could be either differential pair or a Vas transistor.
Here's the rub. The two NPN and PNP transistors need to be matched. Then they need to be matched together. So that is 4 transistors with the same hFE. If you don't have a way of matching transistors from a group, you may want to have it serviced by a good tech. You will need to buy a bunch of transistors to select from. ECG and NTE parts are no flyers. If you attempt to use those, you have lost the right to own any audio equipment. TV's are fine.
It is possible to have a noisy resistor. I've seen that before too. Try to keep an open mind.
Now, does the noise get louder as it gets warmer, or the reverse? Does temperature matter? Do you have a digital voltmeter? If so, measure the DC offset across the speaker terminals with the amp warmed up, no signal.
-Chris
Answers to Queries
Hey, thanks for the quick response.
What is the differential stage? Sorry, I did say I was ignorant didn't I?
Class A (VAS) transistor(s). Are those the round germanium-looking can transistors with the heatsinks glued on?
Feedback shunt capacitor? The only caps I haven't replaced are the giant power supply caps and two non-polar electrolytics, 470 mf 6.3V. Still waiting for the 470s to arrive, won't bother with the big boys. Can you tell me which cap that would be? I've attached the schematic and the parts list is at http://users.otenet.gr/~athsam/HAFLER_DH_200.htm.
The noise is generally a steady hiss, but there are TINY intermittent fizzes and crackles too. The volume is NOT the same -- the noisy channel is marginally less loud than the other. I'd have to say the noise gets quieter as the amp stays on, but I'll have to give it a careful listen tomorrow.
I feel silly obsessing over this. The noise is minor and invisible when the music's playing, but now that I've heard it I can't stop hearing it. The amp is very nice other than this issue, very open and detailed. FYI I'm running this with an old rebuilt Knight tube pre and a 10 year old NAD CD player.
I've had some bad experiences with ECG and NTE transistors early on. No more I swear.
Hey, thanks for the quick response.
What is the differential stage? Sorry, I did say I was ignorant didn't I?
Class A (VAS) transistor(s). Are those the round germanium-looking can transistors with the heatsinks glued on?
Feedback shunt capacitor? The only caps I haven't replaced are the giant power supply caps and two non-polar electrolytics, 470 mf 6.3V. Still waiting for the 470s to arrive, won't bother with the big boys. Can you tell me which cap that would be? I've attached the schematic and the parts list is at http://users.otenet.gr/~athsam/HAFLER_DH_200.htm.
The noise is generally a steady hiss, but there are TINY intermittent fizzes and crackles too. The volume is NOT the same -- the noisy channel is marginally less loud than the other. I'd have to say the noise gets quieter as the amp stays on, but I'll have to give it a careful listen tomorrow.
I feel silly obsessing over this. The noise is minor and invisible when the music's playing, but now that I've heard it I can't stop hearing it. The amp is very nice other than this issue, very open and detailed. FYI I'm running this with an old rebuilt Knight tube pre and a 10 year old NAD CD player.
I've had some bad experiences with ECG and NTE transistors early on. No more I swear.
Attachments
Hey leolabs, You can still download the service manual from the Hafler site. I just did AND they left the schematic in tact inside.
Neddyboy, The differential stage is four transistors Q1, Q2, Q5, and Q6. Anatech advised matching the gain for all four diff transistors. He's so right. You can go around chasing your tail if you don't take the time to do this properly. ....first a positive DC offset of 100mV, then a negative offset of 118mV. ....and it's a pain to remove the board (while losing the fiber washers each time. Now the good news. Some cheap voltmeters have an Hfe test built-in. And weirdly enough, it seems to be the more cheap (I mean inexpensive) the meter, the more likely it will have this Hfe measurement feature. I would never publish any Hfe readings these meters make as gospel, but for a relative measurement they're useful. That's what you're looking for, a real similarity in Hfe for the devices that you put in. Just get'm close.
The VAS's (this one has complementary VAS's), or Voltage Amplification Stage transistors are Q8 and Q11. They are TO-5's with the hats (heatsinks). I think there are maybe a couple more heatsunk transistors on the board, but my memory is concentrating on beer. The number's on those guy's are: 2N3440 (Q11) and 2N5401 (Q7). The numbers will be obscurred by heatsink and brown paint.
The feedback shunt capacitor C5 (470uF/ 6.3V N.P.) ...ehhh, maybe.
Anatech said something about a resistor flaking out. As I look at the schematic, I have a memory of that too. I think it was somewhere in the differential-land and it could be found with freeze spray. Sorry I don't remember which one.
Neddyboy, The differential stage is four transistors Q1, Q2, Q5, and Q6. Anatech advised matching the gain for all four diff transistors. He's so right. You can go around chasing your tail if you don't take the time to do this properly. ....first a positive DC offset of 100mV, then a negative offset of 118mV. ....and it's a pain to remove the board (while losing the fiber washers each time. Now the good news. Some cheap voltmeters have an Hfe test built-in. And weirdly enough, it seems to be the more cheap (I mean inexpensive) the meter, the more likely it will have this Hfe measurement feature. I would never publish any Hfe readings these meters make as gospel, but for a relative measurement they're useful. That's what you're looking for, a real similarity in Hfe for the devices that you put in. Just get'm close.
The VAS's (this one has complementary VAS's), or Voltage Amplification Stage transistors are Q8 and Q11. They are TO-5's with the hats (heatsinks). I think there are maybe a couple more heatsunk transistors on the board, but my memory is concentrating on beer. The number's on those guy's are: 2N3440 (Q11) and 2N5401 (Q7). The numbers will be obscurred by heatsink and brown paint.
The feedback shunt capacitor C5 (470uF/ 6.3V N.P.) ...ehhh, maybe.
Anatech said something about a resistor flaking out. As I look at the schematic, I have a memory of that too. I think it was somewhere in the differential-land and it could be found with freeze spray. Sorry I don't remember which one.
neddyboy,
It would be good if you borrowed a Variac so you could bring up voltages slowly when troubleshooting. You don't want to let the magic smoke out of any parts.
A noisy component (not hum) could be anything. With the noisy channel turned on with a speaker connected -- and no input -- you could try a couple of things and listen for a difference in the noise:
1. Tap parts on the PCB with a plastic handle of a toothbrush. Perhaps you will find that tapping one part changes the amount of noise.
2. Get a can of Freeze-It and apply its cold spray selectively to various parts and listen for a change in the amount of noise.
What is the serial number of your amp? What is the color of your PCBs? Early versions of this amp had phenolic PCBs and their traces were glued on and the assembly solder bath could not operate at optimum temperature. If your PCBs are brown in color you could have a cold solder joint. Get at the back side of the PCB and reflow all the solder joints, especially those for the transistors with the metal tops.
It would be good if you borrowed a Variac so you could bring up voltages slowly when troubleshooting. You don't want to let the magic smoke out of any parts.
A noisy component (not hum) could be anything. With the noisy channel turned on with a speaker connected -- and no input -- you could try a couple of things and listen for a difference in the noise:
1. Tap parts on the PCB with a plastic handle of a toothbrush. Perhaps you will find that tapping one part changes the amount of noise.
2. Get a can of Freeze-It and apply its cold spray selectively to various parts and listen for a change in the amount of noise.
What is the serial number of your amp? What is the color of your PCBs? Early versions of this amp had phenolic PCBs and their traces were glued on and the assembly solder bath could not operate at optimum temperature. If your PCBs are brown in color you could have a cold solder joint. Get at the back side of the PCB and reflow all the solder joints, especially those for the transistors with the metal tops.
Hi neddyboy,
The best transistor checker for gain is going to be something along the lines of the analog Heathkit IT-18 or similar. I get horrible matches from the digital models.
You don't care what the gain is as long as they are all the same. Preselecting these to be close and final selection with an analog unit works the fastest. Just make the pointer go to the same cal mark without having to adjust any knobs except the polarity switch. You can rig something up to do this as well.
Dick is referring to the old phenolic PCB material (brown). These tend to warp and crack traces sometimes.
What are the DC offset voltages on each channel before we get ahead of ourselves? If you see the offset moving around a lot you may have a bad 470 uF cap. Use at least 16V units to replace those with. A bad transistor in one of the diff pairs will do this too.
These amplifiers benefit from matching the transistors. It actually improves the sound quality.
-Chris
The best transistor checker for gain is going to be something along the lines of the analog Heathkit IT-18 or similar. I get horrible matches from the digital models.
You don't care what the gain is as long as they are all the same. Preselecting these to be close and final selection with an analog unit works the fastest. Just make the pointer go to the same cal mark without having to adjust any knobs except the polarity switch. You can rig something up to do this as well.
Dick is referring to the old phenolic PCB material (brown). These tend to warp and crack traces sometimes.
What are the DC offset voltages on each channel before we get ahead of ourselves? If you see the offset moving around a lot you may have a bad 470 uF cap. Use at least 16V units to replace those with. A bad transistor in one of the diff pairs will do this too.
These amplifiers benefit from matching the transistors. It actually improves the sound quality.
-Chris
Work Cut Out For Me
Great stuff gentlemen! I have the version with the light brown PCB boards, so I'm going to reflow all the solder. When the 470mf caps come in I'll get those installed, then I'll see if I need to mess with the transistors.
Thanks again, and I'll keep you posted when I have some results.
Great stuff gentlemen! I have the version with the light brown PCB boards, so I'm going to reflow all the solder. When the 470mf caps come in I'll get those installed, then I'll see if I need to mess with the transistors.
Thanks again, and I'll keep you posted when I have some results.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.