This is somewhat me circling back to an issue I was having months ago. I've been working on a Marantz 1200 on and off for a quite some time. I'm convinced its possessed by the previous owner at this point...
The issue:
-Ground loop-ish hum on the right channel only (next to the transformer)
-Only when the selector is set to "System 1" or "System 2" (Not when set to "Both")
Remediation:
-Changed the filter caps as some have suggested
-Changed the output transistors (in case it was actually oscillation)
-Biased to 14 mv across bases
-DC offset in sub mv range
Notes:
-Was a 120Hz (mostly) square wave with some higher frequency noise on top.
-Bias adjustment did have some influence on the intensity of the hum
-Adjusting bias on left channel has significant influence on hum in right channel (Why??)
-When in my garage, Simply connecting my DMM to the base leads was able to eliminate the hum (prior to transistor swap)
-It is now only occurring in my room, not in my garage (makes it real fun to troubleshoot)
-Prior to filter cap swap, it was always audible through headphones. (weather set to Phones only/System 1/System 2/Both)
-Prior to the transistor swap, it was occurring in my garage, AND in my room
-Connecting the common ground screw on the back of the amp to earth ground in my room does NOT influence the hum (I thought it might)
I'm at a loss, and any help would be appreciated... Thanks yall!
The issue:
-Ground loop-ish hum on the right channel only (next to the transformer)
-Only when the selector is set to "System 1" or "System 2" (Not when set to "Both")
Remediation:
-Changed the filter caps as some have suggested
-Changed the output transistors (in case it was actually oscillation)
-Biased to 14 mv across bases
-DC offset in sub mv range
Notes:
-Was a 120Hz (mostly) square wave with some higher frequency noise on top.
-Bias adjustment did have some influence on the intensity of the hum
-Adjusting bias on left channel has significant influence on hum in right channel (Why??)
-When in my garage, Simply connecting my DMM to the base leads was able to eliminate the hum (prior to transistor swap)
-It is now only occurring in my room, not in my garage (makes it real fun to troubleshoot)
-Prior to filter cap swap, it was always audible through headphones. (weather set to Phones only/System 1/System 2/Both)
-Prior to the transistor swap, it was occurring in my garage, AND in my room
-Connecting the common ground screw on the back of the amp to earth ground in my room does NOT influence the hum (I thought it might)
I'm at a loss, and any help would be appreciated... Thanks yall!
Welcome to the forum!
Is hum affected by the volume control, or by any other controls?
"-Was a 120Hz (mostly) square wave with some higher frequency noise on top." Do you have a scope?
I assume the System 1/ System 2/ Both is a speaker selection switch? Does this switch have any additional function other than speakers? Is hum in headphones affected by speaker selection? No hum in "Both" position? Are there two speakers connected? What if you unhook a speaker when set to Both?
"-Biased to 14 mv across bases". Would you elaborate measurement details? I wonder if meter is coupling ambient noise or stray capaciatance into amp during measurement.
You might monitor ripple on the filter caps while you experiment with things that affect hum amplitude: if you don't have a scope, set your DVM to AC and probe caps for ripple; probably a good precaution to have a 0.1uF cap in series with meter to ensure DC on the supply is rejected by the meter.
Is hum affected by the volume control, or by any other controls?
"-Was a 120Hz (mostly) square wave with some higher frequency noise on top." Do you have a scope?
I assume the System 1/ System 2/ Both is a speaker selection switch? Does this switch have any additional function other than speakers? Is hum in headphones affected by speaker selection? No hum in "Both" position? Are there two speakers connected? What if you unhook a speaker when set to Both?
"-Biased to 14 mv across bases". Would you elaborate measurement details? I wonder if meter is coupling ambient noise or stray capaciatance into amp during measurement.
You might monitor ripple on the filter caps while you experiment with things that affect hum amplitude: if you don't have a scope, set your DVM to AC and probe caps for ripple; probably a good precaution to have a 0.1uF cap in series with meter to ensure DC on the supply is rejected by the meter.
Thank you for the welcome. I have a few other projects on the horizon I'll likely be needing help on.
Its unaffected by the volume knob, and is present even with the preamp -> main-in jumpers
This is correct. my understanding is that is simply provides a path for the output signal to pass through. (Seems dangerous having up to 400 watts pass through a switch but hey, the 70's were a wild time) In the "Both" position, I only have one set of speakers hooked up to either system 1/system 2 (interchanging terminals has no effect)
So the service manual for this model suggests using a Variac to perform the bias adjustment, but alas, I do not have one. On another forum, it was suggested to set bias across the base of the output transistors to 11-14 mv. I could connect my meter directly to the external case of the transistors, but i dont have big enough alligator clips, or coordination, so I used the grabber leads on the base opposing bases. 7mv per base with respect to ground, or 14 mv across both.
My immediate thought was the capacitance of the meter itself, but I don't want to perform modifications unless absolutely required.
I don't have a recent reading on the ripple across the caps, but last I checked, I believe it was 20mv PkPk, or 20mv RMS. I don't recall, but the filters are new 22000 uf 63 volt kemets. I do have a scope, but have yet to haul it to my room. Heres an old picture of what it was doing before on the speaker terminals. Im assuming its the same, because it sounds exactly the same to me. Don't trust the voltage/scale because I may have had the x10 switch on my scope engaged in one channel.
My next step is going to be replacing the rectifier diodes and the bias circuit diodes. I'm leaning towards bias diodes since its mostly in one channel, but i may as well order everything at once to avoid more shipping costs. Shipping will be more expensive than all the non transistor silicon on the whole device.
Its unaffected by the volume knob, and is present even with the preamp -> main-in jumpers
This is correct. my understanding is that is simply provides a path for the output signal to pass through. (Seems dangerous having up to 400 watts pass through a switch but hey, the 70's were a wild time) In the "Both" position, I only have one set of speakers hooked up to either system 1/system 2 (interchanging terminals has no effect)
So the service manual for this model suggests using a Variac to perform the bias adjustment, but alas, I do not have one. On another forum, it was suggested to set bias across the base of the output transistors to 11-14 mv. I could connect my meter directly to the external case of the transistors, but i dont have big enough alligator clips, or coordination, so I used the grabber leads on the base opposing bases. 7mv per base with respect to ground, or 14 mv across both.
My immediate thought was the capacitance of the meter itself, but I don't want to perform modifications unless absolutely required.
I don't have a recent reading on the ripple across the caps, but last I checked, I believe it was 20mv PkPk, or 20mv RMS. I don't recall, but the filters are new 22000 uf 63 volt kemets. I do have a scope, but have yet to haul it to my room. Heres an old picture of what it was doing before on the speaker terminals. Im assuming its the same, because it sounds exactly the same to me. Don't trust the voltage/scale because I may have had the x10 switch on my scope engaged in one channel.
My next step is going to be replacing the rectifier diodes and the bias circuit diodes. I'm leaning towards bias diodes since its mostly in one channel, but i may as well order everything at once to avoid more shipping costs. Shipping will be more expensive than all the non transistor silicon on the whole device.
** Its too late to go back and edit my original reply, but it's present with or without the jumpers in. It does somewhat effect the tone sound, but its still very much there.
Also, from looking at the output, I'm 90% sure I had x10 set on the scope for the right/green channel. 200 mv scale would be accurate, and its amplitude is higher because the right side was noticeably louder.
Hi,
Some scatter shot thoughts:
Are all mounting screws that have electrical bonding to chassis clean and tight?
This is very strange, as the selector switch controls only speaker selection and you have only one speaker connected. I presume you can monitor with your scope and with your headphones. Is hum present in 'phones? How do things change if you move speaker to other binding posts? What happens if you set to "Both" but speakers are disconnected--- hum in 'phones? I imagine that with Both selected, there is more internal wiring in play within the chassis and more stray coupling--- hard to imagine this is relevant, but I mention anyway.
Have you seen any evidence of oscillation? The "grass" the scope channel 2 looks suspicious, as do some of the odd interactions you describe.
BTW, I recommend always using x10 scope probes unless you are desperate for sensitivity. The load capacitance in x1 probes can provoke oscillation when probing internal circuit nodes.
Would you identify what bases you are monitoring when you adjust bias to 14mV? And this is done with no load applied?
I don't know why hum amplitude would be affected by bias adjustment but it's concerning. It suggests that supply ripple may be impacted, but supply ripple should be rejected anyway. As a diagnostic, what if you set bias to minimum--- what happens to hum phenomena? Anything affect the grass in the scope trace?
Good luck!
Some scatter shot thoughts:
Are all mounting screws that have electrical bonding to chassis clean and tight?
This is very strange, as the selector switch controls only speaker selection and you have only one speaker connected. I presume you can monitor with your scope and with your headphones. Is hum present in 'phones? How do things change if you move speaker to other binding posts? What happens if you set to "Both" but speakers are disconnected--- hum in 'phones? I imagine that with Both selected, there is more internal wiring in play within the chassis and more stray coupling--- hard to imagine this is relevant, but I mention anyway.
Have you seen any evidence of oscillation? The "grass" the scope channel 2 looks suspicious, as do some of the odd interactions you describe.
BTW, I recommend always using x10 scope probes unless you are desperate for sensitivity. The load capacitance in x1 probes can provoke oscillation when probing internal circuit nodes.
Would you identify what bases you are monitoring when you adjust bias to 14mV? And this is done with no load applied?
I don't know why hum amplitude would be affected by bias adjustment but it's concerning. It suggests that supply ripple may be impacted, but supply ripple should be rejected anyway. As a diagnostic, what if you set bias to minimum--- what happens to hum phenomena? Anything affect the grass in the scope trace?
Good luck!
Well, I'm done done with my spring semester, so I'm back to working on the 1200. It is a 3500+ serial, so the 1200 addendum is the only correct schematic for this unit, and to some degree, some of the 250.
I pulled all the suspect transistors, and a few were way off hFE, but they have since been changed. Bias was set to 14mv, and again at +10w per channel with my AC wattage meter, but the issue still persists. the bias isnt "even" on both sides relative to ground, but they are even relative to the speaker out positive wire, to account for slight DC offset/drift.
The hum does go away when the bias is all the way down, but it almost sounds like there is some artifact present in the audio quality with the bias down to zero. It also goes away if the lid is removed now. I have a video if no one believes me, but the closer the lid gets, the louder the buzz gets, starting around half a foot away from chassis
It also goes away when the pre-out/main-in plugs don't connect ground. There is a 28 ohm resistor parallel with a capacitor i cant recall on value between the main-in ground, and chassis. This is bypassed when the input plugs are connected, so there is something actually changing value in this case.
The only voltage reading I've found so far that's simi out of spec is the base of Q502. its reading -470 mv and change, but it should be -300 mv. Not the worst ive seen, but that is the differential pair. They have been replaced with matching KSC1845 and thermally/mechanically bonded, yet no effect.
Attachments
There is thw AC connections right next to the bias transostor for the right channel. Once i swap boards, if the issue persists, I'll add a ground shield aroud it. If it doesnt, I'll set a day to pull and test every component on the faulty side. Over on AK, someone fixed their hum by replacing a faulty carbon 1.2k resistor on the dc rail, but they are fine on this board. They arent wirewound either, so i wouldnt expect inductance to be a huge issue.
The main filters have been replaced, but im not seeing any anomalies with the voltage ripple, but I can post the waveform tonight. It was biased per the manual to consume 7-10 watts more per channel relative to no bias, so i'll have to do some math to quatify if the PS is behaving as expected. The rectifier diodes are original, but I didnt observe any signs of leakage, so i left them that way.
I've attached a service manual for convenience to members wishing to contribute.
I recommend against wholesale pulling/testing of components--- too easy to damage board or components in the process, and the problem sounds like a hum/grounding issue, based upon your comments:
Is hum present with volume control at 0? With or without input plugs connected?
I would try to set bias current to about nominal even though hum increases. Zero bias may explain the distortion you alluded to, and larger hum signal may be easier to troubleshoot. CR501 and C503 form sub regulation. Are they intact with low ripple?
I recommend against wholesale pulling/testing of components--- too easy to damage board or components in the process, and the problem sounds like a hum/grounding issue, based upon your comments:
Is hum present with volume control at 0? With or without input plugs connected?
I would try to set bias current to about nominal even though hum increases. Zero bias may explain the distortion you alluded to, and larger hum signal may be easier to troubleshoot. CR501 and C503 form sub regulation. Are they intact with low ripple?