The original stock input caps on this amp are actually 16v 220uF Silmics, although the phono stage output signal goes through 2.2uF first. I want to preserve the essence of the Marantz sound that I enjoy but improve the detail and stereo imaging while delivering a controlled, accurate bass as deep as my speakers want to go. Other than following Brent's suggested changes, I'm mostly just replacing stock parts with better quality or closer matched ones of the same value, but I'm feeling confident that there's some room for me to experiment with the coupling caps. I don't have a scope to prove the response doesn't roll off too soon, but I'll play some test sweeps and listen to my favourite bass lines.
I'm back in action, guys. All mods re-done on a fresh board and it's sounding great.
A couple of things I learned this time: If you're planning to put mount large off board PSU caps in this amp then Kimber 4TC seems to be the perfect wire for the job. It's just the right thickness to fit through the holes left by the snap-in Elna caps. Also, if you're putting larger caps in, I found that it's best to start by cutting down all those resistors and mounting them flush on the PCB surface so you can keep the new component leads as short as possible. I also shielded my pre-amp chip before it got too crowded in there.
A couple of things I learned this time: If you're planning to put mount large off board PSU caps in this amp then Kimber 4TC seems to be the perfect wire for the job. It's just the right thickness to fit through the holes left by the snap-in Elna caps. Also, if you're putting larger caps in, I found that it's best to start by cutting down all those resistors and mounting them flush on the PCB surface so you can keep the new component leads as short as possible. I also shielded my pre-amp chip before it got too crowded in there.
Latest shot of the amp, now with 3.3uF Mundorf Silver Oils inserted between the input selection switch and volume pot. They are a massive improvement over the Black Gate AC, which I have replaced with wire jumpers. After redoing all mods from scratch over two days, I reckon they are one of the three most significant improvements, along with the BG FK smoothers and the extra PSU.
I hesitate to say this but I think I'm close to being done. I'm so happy with the sound right now, I've even put the screws in
R806 is a 1k2 1/4 Watt, you might want to check R804 as well 47R 1/4 Watt fusible.
PM-66 KI blowing fuse
Hi guys.
Can't believe the legs this post has ;-) Impressive work going on here.
My question, at least at this stage, is a little more mundane. A 66 KIS I picked up yesterday blows its 1.6amp internal fuse a lot of the time. I bought it cheapish (£25), as it was described as being dead as a doornail.
A couple of questions arise: is the 1.6A fuse slow- or quick-blow? Given the in-rush on a toroid, I imagine it's slow-blow, and the fuses I'm using at the moment might be quick-blow (have to check). Are the 12,000uF blue Elna reservoir caps attached to the PCB with that adhesive that quite a few manufacturers seem to use? Around the base, for about half the circumference of each cap, is some brown stuff. Wanted to check if its adhesive or leakage. Not getting any major offset on either output when the fuse doesn't blow (10-20mV).
How prone are these amps to bias drift over the years? I have another 66 KIS which seems to work fine, but I remember it sounding better (I owned one years ago, when they first came out). It's in that classic system of the KIS amp and CDP with Tannoy Mercury M1s. Got the CDP cheapish too (£30), as it squeals when loading or unloading a disc. Any pointers there would be appreciated too ;-)
Once one of my amps is fully sorted, the modding will begin!
Cheers, Jon.
Hi guys.
Can't believe the legs this post has ;-) Impressive work going on here.
My question, at least at this stage, is a little more mundane. A 66 KIS I picked up yesterday blows its 1.6amp internal fuse a lot of the time. I bought it cheapish (£25), as it was described as being dead as a doornail.
A couple of questions arise: is the 1.6A fuse slow- or quick-blow? Given the in-rush on a toroid, I imagine it's slow-blow, and the fuses I'm using at the moment might be quick-blow (have to check). Are the 12,000uF blue Elna reservoir caps attached to the PCB with that adhesive that quite a few manufacturers seem to use? Around the base, for about half the circumference of each cap, is some brown stuff. Wanted to check if its adhesive or leakage. Not getting any major offset on either output when the fuse doesn't blow (10-20mV).
How prone are these amps to bias drift over the years? I have another 66 KIS which seems to work fine, but I remember it sounding better (I owned one years ago, when they first came out). It's in that classic system of the KIS amp and CDP with Tannoy Mercury M1s. Got the CDP cheapish too (£30), as it squeals when loading or unloading a disc. Any pointers there would be appreciated too ;-)
Once one of my amps is fully sorted, the modding will begin!
Cheers, Jon.
Hi Triophile,
I'm not sure about the stock fuse, but I'm using a 1.6A slow blow with two transformers and no issues.
Yes, the brown stuff is just glue. It goes quite manky-looking over time.
Whether the bias naturally drifts or not, the pots are very cheap and easily knocked out of place, so I think the best thing to do is just measure it as per the guide in the service manual. I recommend using crocodile clips to save you having to hold the DMM still for several minutes.
Try taking the lid off the CDP and you might be able to tell better why it's squeaking. It might want a bit of white grease on the gears.
Cheers,
Ben
I'm not sure about the stock fuse, but I'm using a 1.6A slow blow with two transformers and no issues.
Yes, the brown stuff is just glue. It goes quite manky-looking over time.
Whether the bias naturally drifts or not, the pots are very cheap and easily knocked out of place, so I think the best thing to do is just measure it as per the guide in the service manual. I recommend using crocodile clips to save you having to hold the DMM still for several minutes.
Try taking the lid off the CDP and you might be able to tell better why it's squeaking. It might want a bit of white grease on the gears.
Cheers,
Ben
Thanks for the reply, Ben.
Mmmm, the plot thickens. The seller was obviously being a little disingenuous when he claimed this PM-66 KIS simply stopped working one day. I finally flipped it over to find that someone has clearly had a go at it internally.
I'm not sure if someone's tried washing the underside of the main PCB, but it looks a bit dirty:
There's also been some reworking, as the solder appears to have been sucked/wicked off one of the pads, and one of the thermal tracking trasistors looks as if its either been reflowed, resoldered, or replaced:
There's also scorching near some of the larger, un-heatsinked transistors:
I'm wondering if someone's tweaked this amp in the past and maybe cooked it slightly, or if perhaps they've biased it into Class A without putting heatsinks on the relevant transistors.
The upshot is that the amp is still blowing fuses, whether slow- or quick-blow 1.6A types. Thing is, it won't blow the first time it's powered up with a fresh fuse, only the second time. Any other suggestions, Ben or anyone else, before I go ahead and check the bias (which might necessitate the purchase of a new DMM, as I barbequed my decent one building a valve amp a while ago)?
Cheers, Jon.
Mmmm, the plot thickens. The seller was obviously being a little disingenuous when he claimed this PM-66 KIS simply stopped working one day. I finally flipped it over to find that someone has clearly had a go at it internally.
I'm not sure if someone's tried washing the underside of the main PCB, but it looks a bit dirty:
An externally hosted image should be here but it was not working when we last tested it.
There's also been some reworking, as the solder appears to have been sucked/wicked off one of the pads, and one of the thermal tracking trasistors looks as if its either been reflowed, resoldered, or replaced:
An externally hosted image should be here but it was not working when we last tested it.
There's also scorching near some of the larger, un-heatsinked transistors:
An externally hosted image should be here but it was not working when we last tested it.
I'm wondering if someone's tweaked this amp in the past and maybe cooked it slightly, or if perhaps they've biased it into Class A without putting heatsinks on the relevant transistors.
The upshot is that the amp is still blowing fuses, whether slow- or quick-blow 1.6A types. Thing is, it won't blow the first time it's powered up with a fresh fuse, only the second time. Any other suggestions, Ben or anyone else, before I go ahead and check the bias (which might necessitate the purchase of a new DMM, as I barbequed my decent one building a valve amp a while ago)?
Cheers, Jon.
Some heat discolouration is normal, but all that white stuff doesn't look so good, especially near the input on the left of the picture. I agree that it looks like someone's worked on it. I wonder if any of that mess is conductive....
I recommend that you get yourself a new multimeter as your first order of business, even before you try powering this on again, as something is definitely not right. You don't want to run the amp with the bias current unknown as you'd only risk damaging it. I'd start by pulling the board and check the dodgy-looking soldering. Some of those joints probably need redoing and there might even be a slight short circuit somewhere.
Once you have checked it and cleaned it up, check and set the bias. The pots should start off at the centre point. I run my amp at 20mV on stock heatsinks.
If you go to an electronics store, pick up a couple of those clip-on transistor heatsinks, as a couple of those run quite hot anyway.
I recommend that you get yourself a one-handed solder sucker as well, if you don't already have one. I find the easiest way to clean up those old joints is to put some more rosin core solder on top to melt whatever's there, suck it all off and then re-solder from clean.
I recommend that you get yourself a new multimeter as your first order of business, even before you try powering this on again, as something is definitely not right. You don't want to run the amp with the bias current unknown as you'd only risk damaging it. I'd start by pulling the board and check the dodgy-looking soldering. Some of those joints probably need redoing and there might even be a slight short circuit somewhere.
Once you have checked it and cleaned it up, check and set the bias. The pots should start off at the centre point. I run my amp at 20mV on stock heatsinks.
If you go to an electronics store, pick up a couple of those clip-on transistor heatsinks, as a couple of those run quite hot anyway.
I recommend that you get yourself a one-handed solder sucker as well, if you don't already have one. I find the easiest way to clean up those old joints is to put some more rosin core solder on top to melt whatever's there, suck it all off and then re-solder from clean.
Last edited:
Sorted, Ben. I'll get all that done.
What I'm hoping to do, once both amps are fine, is split the crossovers in the Tannoys and fit two pairs of binding posts to each, then run them bi-amped, using the volume control on the CDP. Next step, if the Tannoys' crossovers are amenable, is to go active, probably with a basic in-car active crossover first. Once that's done, I'll start modding the amps themselves.
I like the idea, which one previous poster mentioned, of a separate PSU for the front-end IC in the PM-66 KIS. I have a couple of pairs of Trio/Kenwood L-08M monoblocs, and each monobloc has two transformers, one for the input/driver stage, the other for the output stage. Even mine, which aren't properly refurbed and have failing electrolytics, have a superbly open, three-dimensional sound (though presumably their PSRR is unlikely to be as good as the IC in the Marantz).
Another idea might be to do what Kenwood did in the L-09M monoblocs, where the two large mains trannies feed one rail each. Of course, going the whole hog would involve a small supply for the Marantz's front-end, and two toroids wired together, or with one for each of the rails, or one for each of the power stage's channels.
Interesting as well was what another previous poster wrote about E/I vs toroid as far as rejection of mains-borne junk goes: E/Is have a far narrower bandwidth in terms of allowing junk from the primary to get into the secondary. I've also got a low-power R-core sitting in the cupboard at home, which might be worth bolting into one of the amps to power its front-end.
Cheers, Jon.
What I'm hoping to do, once both amps are fine, is split the crossovers in the Tannoys and fit two pairs of binding posts to each, then run them bi-amped, using the volume control on the CDP. Next step, if the Tannoys' crossovers are amenable, is to go active, probably with a basic in-car active crossover first. Once that's done, I'll start modding the amps themselves.
I like the idea, which one previous poster mentioned, of a separate PSU for the front-end IC in the PM-66 KIS. I have a couple of pairs of Trio/Kenwood L-08M monoblocs, and each monobloc has two transformers, one for the input/driver stage, the other for the output stage. Even mine, which aren't properly refurbed and have failing electrolytics, have a superbly open, three-dimensional sound (though presumably their PSRR is unlikely to be as good as the IC in the Marantz).
Another idea might be to do what Kenwood did in the L-09M monoblocs, where the two large mains trannies feed one rail each. Of course, going the whole hog would involve a small supply for the Marantz's front-end, and two toroids wired together, or with one for each of the rails, or one for each of the power stage's channels.
Interesting as well was what another previous poster wrote about E/I vs toroid as far as rejection of mains-borne junk goes: E/Is have a far narrower bandwidth in terms of allowing junk from the primary to get into the secondary. I've also got a low-power R-core sitting in the cupboard at home, which might be worth bolting into one of the amps to power its front-end.
Cheers, Jon.
Hey Jon,
Your speakers are two way monitors, am I right? I'm wondering what benefit you expect to get from bi-amping them? My B&Ws are two-way also, and have two sets of binding posts each for bi-amping, but I thought tweeters take so little power that it wasn't really worth it. I presume if you're planning on doing different mods for each amp then you're thinking of using a whole amplifier just to drive your tweeters rather than one amp per channel?
I don't want to put a downer on your plans just as you're getting started, but I have to give you a heads up about the volume control on the CD player - you'll have to listen to it yourself and make your own mind up, but I personally think it degrades the sound quality quite badly. Before you lose your pot, try comparing the sound of the CDP at full volume versus any other level and I think you'll hear a difference. If you hear what I hear then you might be better using a separate passive pre-amp or find out a way to use the volume pot in one of the amps.
Cheers,
Ben
Your speakers are two way monitors, am I right? I'm wondering what benefit you expect to get from bi-amping them? My B&Ws are two-way also, and have two sets of binding posts each for bi-amping, but I thought tweeters take so little power that it wasn't really worth it. I presume if you're planning on doing different mods for each amp then you're thinking of using a whole amplifier just to drive your tweeters rather than one amp per channel?
I don't want to put a downer on your plans just as you're getting started, but I have to give you a heads up about the volume control on the CD player - you'll have to listen to it yourself and make your own mind up, but I personally think it degrades the sound quality quite badly. Before you lose your pot, try comparing the sound of the CDP at full volume versus any other level and I think you'll hear a difference. If you hear what I hear then you might be better using a separate passive pre-amp or find out a way to use the volume pot in one of the amps.
Cheers,
Ben
Hi Ben.
I've bi-amped before, and it's always brought improvements, sometimes very obvious, sometimes subtler. Don't forget, the amps aren't driving the tweeters or mid/bass units directly - they have to go through the crossovers, which, in conjunction with the drivers, tend to be relatively tough loads compared to a simple, purely resistive load. Separating the amps also reduces intermodulation distortion, AFAIA: ie, where the big bass peaks modulate, to a certain extent, the more delicate high-frequency signals.
In terms of arranging the amps, I was going to try one per speakers as well as the alternative, one per pair of the same driver type, to see which works best, Anyway, all of this is only a stepping stone towards going active, which, if done properly, invariably yields very clear benefits.
I'll have another listen to the CDP's volume control (not something I've done in any critical way so far, as the system as it currently stands is in my girlfriend's living room). I could always keep an eye out for a pre, and tweak that too (I'm thinking stepped attenuator, certainly).
Cheers, Jon.
I've bi-amped before, and it's always brought improvements, sometimes very obvious, sometimes subtler. Don't forget, the amps aren't driving the tweeters or mid/bass units directly - they have to go through the crossovers, which, in conjunction with the drivers, tend to be relatively tough loads compared to a simple, purely resistive load. Separating the amps also reduces intermodulation distortion, AFAIA: ie, where the big bass peaks modulate, to a certain extent, the more delicate high-frequency signals.
In terms of arranging the amps, I was going to try one per speakers as well as the alternative, one per pair of the same driver type, to see which works best, Anyway, all of this is only a stepping stone towards going active, which, if done properly, invariably yields very clear benefits.
I'll have another listen to the CDP's volume control (not something I've done in any critical way so far, as the system as it currently stands is in my girlfriend's living room). I could always keep an eye out for a pre, and tweak that too (I'm thinking stepped attenuator, certainly).
Cheers, Jon.
Hey Jon, thanks for the explanation. I just assumed most of the power of the tweeter amp would end up getting dissipated as heat.
A stepped attenuator sounds like a neat way to get two amps to work together in tandem. I've seen some DACT ones that have several channels stacked together.
Cheers,
Ben
A stepped attenuator sounds like a neat way to get two amps to work together in tandem. I've seen some DACT ones that have several channels stacked together.
Cheers,
Ben
Hi folks,
I'm concerned about DC offset. Can anyone help me out?
Firstly, I measured across the output of my amp with the speakers connected and no music playing. Is that a valid test?
Secondly, assuming I measured correctly, does anyone know how I can trim my DC offset at the output? B&W advised me that it should not exceed 25mV, and I think within 20mV is the generally accepted allowable limit.
Thanks in advance,
Ben
I'm concerned about DC offset. Can anyone help me out?
Firstly, I measured across the output of my amp with the speakers connected and no music playing. Is that a valid test?
Secondly, assuming I measured correctly, does anyone know how I can trim my DC offset at the output? B&W advised me that it should not exceed 25mV, and I think within 20mV is the generally accepted allowable limit.
Thanks in advance,
Ben
Hey guys, I finally got around to doing the Schottky bridge. Wow, what a great mod! It's made a huge improvement! I've had the diodes just sitting there for more than a year.
I also have the beefy-looking 100 watt transistors sitting in my box of bits. Any opinions on whether I should attempt swapping those in too? I don't think they're the same manufacturer - pos are Tosh and neg are Magnatec or something - don't know if that matters at all.
I also have the beefy-looking 100 watt transistors sitting in my box of bits. Any opinions on whether I should attempt swapping those in too? I don't think they're the same manufacturer - pos are Tosh and neg are Magnatec or something - don't know if that matters at all.
Hi Ben,
My best guess is that it does matter a great deal. It has been said that transistors are the single most important component in terms of quality vs the effect on sound. Even if that effect is being exaggerated I would make sure they're all good ones because they'll be quite a fuss to change.
Simon
My best guess is that it does matter a great deal. It has been said that transistors are the single most important component in terms of quality vs the effect on sound. Even if that effect is being exaggerated I would make sure they're all good ones because they'll be quite a fuss to change.
Simon
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.