Marantz 4230 right power amp board is missing H705 (2SC1318Q)

[vferdman]

Update.

I have replaced the bad 2SA684 transistor with KSA1220, but found more bad stuff. A 2SA641 PNP transistor is bad on one side. It is paired with 2SC945 NPN. I replaced it with a 733 PNP. Not sure if it will match up to the 945 properly. I also found a bad varistor on that board. There are two of them and one was open circuit. Fortunately I was able to find an old varistor from a 2270 restore (I think it's a different part, but I was able to get the biasing circuit to work in a stable manner and get it to adjust to 15mV as per manual). The other varistor was good, but the leads broke off flush with the package and I had to do surgery to open up the package enough to solder the leads back on and secure it with epoxy. Now I have a stable amp board, but the signal gets distorted somewhere along the line. I get very early clipping (compared to the good left channel) and the sine wave does not look right at anything but very low volume. I will now have to trace the singanal through and see what else is bad. So far I replaced emitter resistors, outputs, some of the intermediate transistors that were bad and all the electrolytic caps, of course.

This thing is really not cooperating and taking a lot of time to repair. Thanks again to all who help and support me. It is a learning experience for sure.

 

[rBuckner]

Update.

I have replaced the bad 2SA684 transistor with KSA1220, but found more bad stuff. A 2SA641 PNP transistor is bad on one side. It is paired with 2SC945 NPN. I replaced it with a 733 PNP. Not sure if it will match up to the 945 properly. I also found a bad varistor on that board. There are two of them and one was open circuit. Fortunately I was able to find an old varistor from a 2270 restore (I think it's a different part, but I was able to get the biasing circuit to work in a stable manner and get it to adjust to 15mV as per manual). The other varistor was good, but the leads broke off flush with the package and I had to do surgery to open up the package enough to solder the leads back on and secure it with epoxy. Now I have a stable amp board, but the signal gets distorted somewhere along the line. I get very early clipping (compared to the good left channel) and the sine wave does not look right at anything but very low volume. I will now have to trace the singanal through and see what else is bad. So far I replaced emitter resistors, outputs, some of the intermediate transistors that were bad and all the electrolytic caps, of course.

This thing is really not cooperating and taking a lot of time to repair. Thanks again to all who help and support me. It is a learning experience for sure.

Let me give an accurate hindsight! I always check EVERY last semi on an amp when I find one bad. Heck, even just troubleshooting!

 

[vferdman]

Update.

It looks like I have a working amp now. I have ended up with all new outputs (TIP41) and most new transistors and diodes. The varistors, as I mentioned earlier are: one original and the other from a 2270 amp I had removed before and have packaged it into the boat shaped package of the original varistor on this 4230 using epoxy and some surgery. It looks like I am getting slightly less power out of the unit than 30 watts before clipping begins. I was able to adjust everything according to the manual, but my 1kHz signal only reaches about 24 watts before it clips. It sounds good and there is no visible distortion, so I am going to cautiously call this complete. I am having the unit powered up overnight with volume at the minimum just to idle it. I will do some more measurements on it for power on both 30W x 2 and 12W x 4 and will report back. The music does sound very good and through it now. Like a Marantz should. Still, I would rather have just the good old 2 channel stereo receiver without all that Dolby nonsense. etc.

Thanks again for your support!

 

[rBuckner]

Great! I agree on the 4 channel / 2 ch thing.

 

[vferdman]

I still have some minor issues with the receiver in 30W x 2 mode. First of all, there is about 300 ohms between common (black) speaker terminals. When I switch to 12W x 4 the resistance goes down to less than 1 ohm. Can anyone explain to me how the unit switches between the two modes? Looking at the schematic I am kind of lost. If someone knows how this is accomplished, please educate me. What I get is distortion on the right channel output when I use a 1kHz waveform into the CD input. However, if I put the probe's ground reference on the left channel black terminal, the output looks good. So I am a bit confused. BTW, my scope ground is floated and is not connected to my home ground. I want to state that before I get asked. So I am referencing my scope to the speaker terminals' black lugs. I am measuring between black and red, just like the speakers are connected. I am seeing a distorted signal on the right, clean signal on the left and if I reference the right channel off the left channel's black terminal, I get a clean output.

Please advise if you can!

 

[vferdman]

Update.

I have found a bad resistor. A 330 ohm resistor R744 was open on the NPN-PNP pair on one half of the right power amp. Strangely enough it worked well enough like that, but there was some distortion at high levels and amp did not reach 30 Watts. Now it works better, getting full 30+ watts before clipping, but there is still some strangeness and there is still no rear right channel in 12W x 4 mode. Still don't understand how the switching between 2 and 4 channel mode works in electronic sense. If anyone can educate me on this, I will be grateful.

 

[pac1085]

Update.

I have found a bad resistor. A 330 ohm resistor R744 was open on the NPN-PNP pair on one half of the right power amp. Strangely enough it worked well enough like that, but there was some distortion at high levels and amp did not reach 30 Watts. Now it works better, getting full 30+ watts before clipping, but there is still some strangeness and there is still no rear right channel in 12W x 4 mode. Still don't understand how the switching between 2 and 4 channel mode works in electronic sense. If anyone can educate me on this, I will be grateful.

im sure someone else can chime in with a more detailed answer, but you'll want to read about 'bridge tied load' - that's how the amp operates when in 2 channel mode. you also have to be very careful connecting it to test equipment when in 2 channel mode. when its in 4 channel mode, it's essentially a true 4 channel amp.

don't connect your scope to both channels when you are running the amp in 2 channel mode, you can't have the common ground of the BNC shell in the bridge tied load mode - when setup like this, the negative terminal is not common between the two channels. you'll burn stuff up. ask me how i know, lol. I'm not an expert by any means but a 4230 was one of the first things I repaired and learned some hard lessons. There are ways around this (like using 4 probes in "pseudo differential" mode, getting actual differential probes, or getting some cheap handheld scopes, or a scope with isolated inputs (rare))

 

[vferdman]

im sure someone else can chime in with a more detailed answer, but you'll want to read about 'bridge tied load' - that's how the amp operates when in 2 channel mode. you also have to be very careful connecting it to test equipment when in 2 channel mode. when its in 4 channel mode, it's essentially a true 4 channel amp.

don't connect your scope to both channels when you are running the amp in 2 channel mode, you can't have the common ground of the BNC shell in the bridge tied load mode - when setup like this, the negative terminal is not common between the two channels. you'll burn stuff up. ask me how i know, lol. I'm not an expert by any means but a 4230 was one of the first things I repaired and learned some hard lessons. There are ways around this (like using 4 probes in "pseudo differential" mode, getting actual differential probes, or getting some cheap handheld scopes, or a scope with isolated inputs (rare))

Thank you so much! I HAVE figured out that there is no common reference between the two channels in 2 channel mode. I measured about 300 ohms of resistance between the two black speaker terminals (with unit unplugged!) in 2 channel mode. When switching to 4 channel mode, there is less than one ohm of resistance between the black terminals on the front set of speaker terminals and the rear terminals have a physical wire connecting the two black terminals. This is definitely a funky setup that I am not familiar with. When measuring in 2 channel mode, I float the ground on my scope's power cord (old HP scope with only two channels) and connect the probe ground of just one probe to the black terminal. Then connect each channel of the scope to each red speaker terminal. This gives funky results for sure. One channel reads fine (the one that has the probe ground on the black speaker terminal). The other is questionable, but at least I can see the waveform and be aware that I can not really see both channels at the same time correctly.

I will look up the bridged load concept for sure. Thanks for the lead. I am still looking for a problem of no right rear output in 4 channel mode. I suspect the switch, but have not found concrete evidence yet. If it's just the switch, then I will leave it alone as the 4 channel mode is not going to be used on this receiver.

Thanks again!

 

[pac1085]

When measuring in 2 channel mode, I float the ground on my scope's power cord (old HP scope with only two channels) and connect the probe ground of just one probe to the black terminal. Then connect each channel of the scope to each red speaker terminal.

don't do that, it’ll burn things up. the probe grounds are tied together on most(practiaclly all) scopes, so even though you aren't connecting the ground from the second probe, it's already connected thru the other one. you can only measure one at a time without special equipment.

 

[vferdman]

don't do that, it’ll burn things up. the probe grounds are tied together on most(practiaclly all) scopes, so even though you aren't connecting the ground from the second probe, it's already connected thru the other one. you can only measure one at a time without special equipment.

I understand how scope probes work. I am not going to burn anything up by connecting two probes with only one of them having a connection to ground (chassis or black speaker terminal). If I connect both ground leads of probes to different potentials, that will cause problems as I will be shorting those two points. Only connecting one ground to anywhere on the receiver, even positive voltage of some sort is ok since my power ground is floating (not connected to my earth ground in the house). What I will get is erroneous reading referencing whatever the scope ground is connected to. So If I hook up two probes with only one probe ground connected (anywhere on the receiver, really) I will get a reading on each channel referenced to that ground connection. Nothing dramatic will happen except I will get bad readings. I have already done this many times. I understand that I can only get a correct reading from one channel at a time unless I do fancy stuff like differential probes.

 

[pac1085]

I understand how scope probes work. I am not going to burn anything up by connecting two probes with only one of them having a connection to ground (chassis or black speaker terminal). If I connect both ground leads of probes to different potentials, that will cause problems as I will be shorting those two points. Only connecting one ground to anywhere on the receiver, even positive voltage of some sort is ok since my power ground is floating (not connected to my earth ground in the house). What I will get is erroneous reading referencing whatever the scope ground is connected to. So If I hook up two probes with only one probe ground connected (anywhere on the receiver, really) I will get a reading on each channel referenced to that ground connection. Nothing dramatic will happen except I will get bad readings. I have already done this many times. I understand that I can only get a correct reading from one channel at a time unless I do fancy stuff like differential probes.

yeah, I see what you're saying. I misread your post a little bit. this gets confusing pretty quick. from what I remember, in 2ch mode, neither black connection is connected to chassis ground, so no point in even attempting to measure more than 1 channel at a time, since you can't connect both blacks to the scope simultaneously. sounds like you are on the right track

 

[vferdman]

yeah, I see what you're saying. I misread your post a little bit. this gets confusing pretty quick. from what I remember, in 2ch mode, neither black connection is connected to chassis ground, so no point in even attempting to measure more than 1 channel at a time, since you can't connect both blacks to the scope simultaneously. sounds like you are on the right track

You are correct. In 2 channel mode neither left nor right speaker is referenced to chassis ground. I measure them by clipping the scope ground to black terminal and reading the red terminal. One at a time. This does work as long as the scope power cord ground is floated. But, yes, it is confusing and this beast is giving me an education, lol.

From what I learned so far about bridge-tied amps, the inputs to either half have to be out of phase. I think (but not sure yet) that first stage in the amp (based on H701) is the unity gain phase inverter that gets hooked up when in 2 channel mode. Not sure, but it's the only part in the schematic that is not symmetric. At this point I get a very good sound out of the receiver in 2 channel mode and good sound out of 3 speakers in 4 channel mode. Right rear is out in 4 channel mode. I am not getting the input to the amp from tracing so far. Trying to figure out where it comes from beyond the 2ch/4ch switch. It may be the switch, which I have examined and cleaned and lubed with deoxit. I am just really confused about where the other side of the switch leads to. I think it's to the tone control, but I am not sure. Trying to figure it out from the schematic is super hard since it's so dense and my eyes are not what they used to be.

 

[vferdman]

Update.

I have traced the dead right rear channel to a place on the tone preamp that I myself have refurbished. Well, I missed a spot where a very short trace got lifted during the transistor and cap replacement. So there needed to be a jumper, which in this case was the length of the lead of said transistor bent to reach the cap's lead (they are close to each other on the board). So the signal unsurprisingly vanished there and did not get to the amp. That took some figuring out, but once I learned where the connections are made on the 4/2 channel switch, I was on the right track straight to the tone board. After fixing this, I have all 4 channels operating properly and I am able to adjust the clipping symmetry and I am getting proper power from the unit in 4 channel and 2 channel modes.

But there is now another problem! The right power amp heats up like crazy when working at moderate power. Significantly more than the left one. I have a very good idea that it is because of the varistor I have replaced on one side of the right power amp. When I let the amp cool down to room temperature, then turn it on and play at normal levels I can feel the part of the heat sink that has the replacement varistor get warm almost immediately if I keep my hand on it. It literally takes seconds before I begin to feel the warmth. Eventually it gets really hot and turning up the volume makes it cut out on the right side. This thing does not have protection circuitry, I don't think (does it?), so I do not let this go on and turn it off (it sounds really good, BTW during this short experiment and with a sine wave it looks perfect). So I am thinking that the varistor I threw in there is not a good choice. It was one I had from a 2270 rebuild, I believe (but not sure). It is a tiny glass ball with a dot on it. It measured a similar value to the original one, so I used the plastic boat shaped package of the bad one, emptied it out and inserted the glass ball into it and put it on. I also had to fabricate the bracket that holds it to the heat sink. The unit came without it (also missing some screws holding the amp board to the heat sink and heat sink to the chassis).

So, my plan is to maybe not use the boat shaped package and mount the glass ball directly to the heat sink so it can react better. I think I have thermally insulated it too much. I would much rather just replace it with the proper part, but I am not sure if I can find one. If anyone here has one they are willing to let go, please do so!

In any case, I think I am very close. There are no more mysteries left except this overheating issue and I am pretty certain the culprit is the varistor. In fact, that may have been the failure originally. The varistor wen causing the whole thing to overheat on that side of the amp. All my dead components were on the side on the bad varistor.

 

[pac1085]

Update.

I have traced the dead right rear channel to a place on the tone preamp that I myself have refurbished. Well, I missed a spot where a very short trace got lifted during the transistor and cap replacement. So there needed to be a jumper, which in this case was the length of the lead of said transistor bent to reach the cap's lead (they are close to each other on the board). So the signal unsurprisingly vanished there and did not get to the amp. That took some figuring out, but once I learned where the connections are made on the 4/2 channel switch, I was on the right track straight to the tone board. After fixing this, I have all 4 channels operating properly and I am able to adjust the clipping symmetry and I am getting proper power from the unit in 4 channel and 2 channel modes.

But there is now another problem! The right power amp heats up like crazy when working at moderate power. Significantly more than the left one. I have a very good idea that it is because of the varistor I have replaced on one side of the right power amp. When I let the amp cool down to room temperature, then turn it on and play at normal levels I can feel the part of the heat sink that has the replacement varistor get warm almost immediately if I keep my hand on it. It literally takes seconds before I begin to feel the warmth. Eventually it gets really hot and turning up the volume makes it cut out on the right side. This thing does not have protection circuitry, I don't think (does it?), so I do not let this go on and turn it off (it sounds really good, BTW during this short experiment and with a sine wave it looks perfect). So I am thinking that the varistor I threw in there is not a good choice. It was one I had from a 2270 rebuild, I believe (but not sure). It is a tiny glass ball with a dot on it. It measured a similar value to the original one, so I used the plastic boat shaped package of the bad one, emptied it out and inserted the glass ball into it and put it on. I also had to fabricate the bracket that holds it to the heat sink. The unit came without it (also missing some screws holding the amp board to the heat sink and heat sink to the chassis).

So, my plan is to maybe not use the boat shaped package and mount the glass ball directly to the heat sink so it can react better. I think I have thermally insulated it too much. I would much rather just replace it with the proper part, but I am not sure if I can find one. If anyone here has one they are willing to let go, please do so!

In any case, I think I am very close. There are no more mysteries left except this overheating issue and I am pretty certain the culprit is the varistor. In fact, that may have been the failure originally. The varistor wen causing the whole thing to overheat on that side of the amp. All my dead components were on the side on the bad varistor.

I hear you on the schematic. I had to print it on a plotter at work, I still have a massive copy of it hanging on my wall as a reminder..lol

The one I worked on had totally destroyed sv-3x (I think is the part number?) in those positions. I wasn’t able to salvage them. My memory is foggy but I know I replaced them with 1n4148s in series, heatshrinked and clamped down. I’m pretty sure I used 2 of them to get around 1.4v drop, which seemed to match one of the good original ones. There seemed to be a lot of confusion online, and I believe some people were recommending using 3 in series, or another part that had a higher drop (sv3x is different than sv03, I think that was the confusion)

 

[vferdman]

I hear you on the schematic. I had to print it on a plotter at work, I still have a massive copy of it hanging on my wall as a reminder..lol

The one I worked on had totally destroyed sv-3x (I think is the part number?) in those positions. I wasn’t able to salvage them. My memory is foggy but I know I replaced them with 1n4148s in series, heatshrinked and clamped down. I’m pretty sure I used 2 of them to get around 1.4v drop, which seemed to match one of the good original ones. There seemed to be a lot of confusion online, and I believe some people were recommending using 3 in series, or another part that had a higher drop (sv3x is different than sv03, I think that was the confusion)Thanks for

I will try two or three diodes in series as well as some other stuff. I believe this is the last problem on this thing. Thank you for the support.

 

[pac1085]

I will try two or three diodes in series as well as some other stuff. I believe this is the last problem on this thing. Thank you for the support.

Found the post I made. Looks like it was 2 diodes.

 

[vferdman]

View attachment 3167802
Found the post I made. Looks like it was 2 diodes.

That is great! Thank you. Can you tell me more about flyback diodes? Looks like a mod worth considering. I have plenty of 1n4148 diodes kicking around to accomplish this. I also did the dremel trick to the other varistor, which was good, but the leads broke while I was handling it and I had to go in and do the dremel thing with solder. I feel it is strong enough. Stronger than the original leads were, so I will leave that one as is unless it gives me problems.

Maybe include a link to the thread where the screen shot is from? I couldn't find it here. Was it on different forum?

 

[pac1085]

That is great! Thank you. Can you tell me more about flyback diodes? Looks like a mod worth considering. I have plenty of 1n4148 diodes kicking around to accomplish this. I also did the dremel trick to the other varistor, which was good, but the leads broke while I was handling it and I had to go in and do the dremel thing with solder. I feel it is strong enough. Stronger than the original leads were, so I will leave that one as is unless it gives me problems.

Maybe include a link to the thread where the screen shot is from? I couldn't find it here. Was it on different forum?

It was on Facebook. If you are a member of the “vintage Marantz conservation and restoration” group, just search for “4230 sv-3x” and my post should come up

I don’t remember who told me about the fly back diodes. I think someone said it was referenced in a service bulletin but I could never find it. There seems to be a lot of talk about it on these forums as well, but not necessarily specific to the 4230. For example: https://audiokarma.org/forums/index.php?threads/of-flyback-diodes-and-bypass-caps.962131/

I don’t believe you’d be able to use a 4148 for that, seems like 1n4004 and up is commonly recommended

 

[vferdman]

It was on Facebook. If you are a member of the “vintage Marantz conservation and restoration” group, just search for “4230 sv-3x” and my post should come up

I don’t remember who told me about the fly back diodes. I think someone said it was referenced in a service bulletin but I could never find it. There seems to be a lot of talk about it on these forums as well, but not necessarily specific to the 4230. For example: https://audiokarma.org/forums/index.php?threads/of-flyback-diodes-and-bypass-caps.962131/

I don’t believe you’d be able to use a 4148 for that, seems like 1n4004 and up is commonly recommended

Thank you. I will look for the FB group. I am not currently signed up for it. I meant 1n4148 diodes for the varistor function, sorry. Flyback are the big ones, 1n4004s. Understood.

 

[vferdman]

Update.

I am almost done with this thing. I have made two identical "varistor" replacements from a pair of 1N4148 diodes and replaced what I had in there, which was an original part on one side and two 1N4148 diodes in series packed into an original boat shaped package. For consistency I have removed both of those (also, the original varistor lost one of its leads again) and replaced both with identical parts I have made up of two 1N4148 in series inside a heat shrink tube with plenty of heatsink compound. This at least made things act symmetrically and at this point I have a fairly well controlled bias situation on both sides of the amp. I am observing with two meters at the same time as I cycle the volume up and down and it settles nicely after I turn the volume down. However, there is another issue. As I turn the volume just up to before clipping and observe my output with the scope (1kHz signal), I see that my two halves of the amp are running at different voltages on the outputs where I measure the bias. Left channel, which is a good one and has not been damaged previously, is symmetrical by comparison. This is all in 2 x 30W mode. I have about 300 mV more on one half of the right amp than the other (would be rear/front in 4 channel mode). The output waveform looks good, though. Still troubleshooting this.

Thanks again to all for the support. This unit is kicking my butt, but I am learning a lot.