Following my recent Amplifier Restorations posts (Arcam Alpha & NAD 3020) comes this Marantz PM7200.

The owner complained that the amp lost one channel and then the other and lastly it refused to come out of protection mode. Was worried that the processor had issues but initial checks showed that the preamp, tone and volume circuitry to be in good order.


With this amp, if the processor dies then you can be pretty sure that nothing on the front panel lights up and the amp will refuse to turn on. So this just leaves the power amp section. Marantz had installed a small access panel at the bottom of the power amp section but you can't test it without removing the board from the chassis so got the rear panel off, disconnected the preamp and power supply cables from it and removed the power amp board from the chassis.


As can be seen from the image above, the power amp uses 8 transistors hidden behind copper shielding. These were tested individually along with the driver transistors and all seemed to be operating well. So proceeded to check the capacitors, first few that were removed were within tolerance though a bit low but then started noticing that some were well below the 20% allowed tolerance.

Took a closer look at the protection circuitry board which is a little thing at the back near the speaker outputs. Most amps have the protection relays or whatever other form of circuitry placed here. As can be seen from the image below, there were 4 caps on this board and all of them tested well below tolerance.


Changed these caps, temporarily installed the power amp board back in the chassis. Powered up the Marantz and the amp came out of protect mode. Ran it for a awhile and noticed that the amp runs a bit hot. Hit the Class A button and it really got hot inside the amp. Things were becoming slightly clearer now. On to the next post...

So I called up the owner and asked how exactly was the Marantz ran in his system. Found out that initially, when the amp was about 2 years old, the same problem occurred but the dealer fixed it FOC. Then it happened again and as he was in the middle of moving and having a child, the Marantz was put away. But when it was in operation, the unit was left on 24/7 and this model (and I'm guessing the PM8000 and PM7000) runs warm even without load. Go into Class A and it gets worse.

This would explain the caps failing after a number of years. Electrolytic caps usually last about 10 years or so with regular operation. This amp was about 6 years old.

I recommended doing a full recap of the unit to the owner and it was agreed upon. Below are some pics:

Input Board Before


Input Board After


Power Amp Board After


Front Faceplate and Tone, Volume and processor board removed


Tone, Volume and Processor Board Removed for Recap


Power Supply Board



Once the amp got recapped, its idling voltage and DC offset was checked. The Marantz has a low idling voltage in AB mode and the voltage goes up significantly when the Class A button is pressed. The heat the unit produces really does take a toll on the capacitors and other components. Some even showed burnt marks.

All caps were changed to Panasonic FM caps and FC capacitors when the FM ones weren't available. All are to original spec as showed in the Marantz PM7200 service manual. There are some slight differences with the PM8000 but internally the boards are the same. The Panasonic caps are good for 105C whereas the original Elna caps the Marantz used were good for 85C.

The main filter caps were still good plus this was done within a budget for the owner and changing the main caps would have cost a bomb.

Listening to the unit running between Class AB and Class A, a small difference in quality was noted but to have the unit run at such high temperatures and to bring down its lifespan, I really cannot see how the Class A mode is justified. Arguably though, the unit does sound a lot better when fully warmed up and this can take about half an hour. This was noted by how the service manual states that the idling current settles down after half an hour and the idling current when cold is only 0.3mV whereas once warmed up and settled down, it does so at 18mV. A big gap.

Also noted was that running it via the processor actually didn't make the sound quality better and using the Source Direct function brought about a better improvement in soundstaging than using the Class A mode.

Hope this helps anyone out there with a Marantz with similar issues, Cheers

This post has been edited by Black Octagon: Apr 11 2013, 11:22 PM

Thanks for sharing...

During Class A mode,idling current is around how much??

Class AB - 18mV
Class A - 90mV

mV? Or is it mA? The original question was about idling current...

Most audio application specific capacitors are rated for 85C, that includes Nichicon MUSE series (such as FG a.k.a "Fine Gold"). I only use them in the audio paths, rarely/hardly in power supply and power amplifier sections. Anyway, besides Panasonic FM and Pansonic FC, there are other common ones like Nichicon PW and Nichicon HE which is well known in DIY audio circles.

Hehe, good eye there...

Was refering to the Marantz service manual's idling current setting which states 18mV in Class AB and 90mV in Class A. I get what you mean about how it should be mA instead of mV but a lot of these service manuals don't actually give the idling current but instead the voltage that it needs to be set to for the transistors current draw.

So these are read across twin 0.33ohm resistors which give a real world reading of 0.8ohm. So idling current in class AB should be 22.5mA and in class A it should be 112.5mA.

I am aware of the Nichicon caps and to use them in signal paths but the point here was to give some higher temp caps as the Marantz puts out a lot of heat in Class A mode and I had most of the Panasonic caps in stock already. Funnily enough, a Creative sub I have in at the moment has 105C caps installed but even then they've all bulged as the Creative puts out a lot of heat.

that didn't sound too good, got any higher rated ones? LOL

If around 112.5mA,then that's Class A at low power(useful listening level),then after reach certain amount of output power,will go back to Class AB...

As for the service manual, methinks you are referring to this part...



Those Panasonic and Nichicons are quality capacitors and pretty robust plus can withstand high temperatures, even though they were not rated to operate at higher temperatures than specifications. As for the Creative subwoofer, are those bulged the original capacitors? What brand were they?

Actually there are capacitors that can operate at higher temperatures, from 125C to 200C. Just that we hardly use them, not considered as a necessity.

hmm 24/7 operations at rather hot temperatures (not sure how hot though) for around 7 years.

maybe its a matter of aging and not so much temperature rating?

Bulging and leaking capacitors are a familiar sight for me. Quite often the quality of the capacitor determines whether it will bulge or not (besides heat and aging effects). I've seen plenty of bulged capacitors in an active cooling environments such as PC power supplies (which has cooling fan nearby). Incidentally some of the ancient PC power supplies which uses good quality capacitors did not bulge.

lex: Yup, that's the part in the service manual.

As for the Creative sub, the caps are mounted right next to the amp's heatsink and there is evidence that it got quite hot inside there judging by the solder joints which have been baked into the board and are a pain to remove.

Gotta take a look at the caps when I get the chance, can't remember what brand they were but considering they lasted 7 years and were right next to the heatsink, guess they weren't that bad and of pretty decent quality.

Lex is right about quality though, think it was during the early to mid 2000's when there was the "capacitor plague" where a lot of manufacturers got a bunch of bad caps from some China capacitor manufacturer which caused equipment to fail prematurely. Bad quality caused leaks and bulges...think Samsung was hit pretty bad with this as they used the caps in the power supplies of their display units.

Recently just serviced a PM7200 and upgraded parts. It was a good amp before but now it's just amazing. Thanks for the guide. I created a new post earlier anyway for my endeavours.

I'm now thinking about some active forced cooling, these Amps run hot, but the output transistors have compensation diodes built in, to regulate bias current with temperature.

Good to hear

But still do think physically they're a bad design, putting the left & right channel heatsinks so close to each other, run it in their "Class A" mode and everything heats up on that amplifier board. Technically the design may be sound but in real world conditions, it isn't built to last and will require work done to it more often than some other amps out there.

Interesting!
I have PMed you for some assistance.

Yes you are right. Actually thinking about it, despite all the reviews in the hi-fi press about 'build' quality of commercial Hi-hi equipment, really they are quite poor. In terms of heat and ageing all the Elna original caps are 1000 hr rated at 85 deg. C. The amp heatsinks probably reach this temp.

Also in reference to the PM7200 build quality it really is a mass produced amp, there are several aspects I would like to improve, but I could end up rebuilding the whole thing! Generally my thoughts:-

1. Inadequate fixings for PCBs such as source, tone and volume boards. After servicing the amp and taking it apart, there really should be more standoffs, or some foam or something to restrain PCB movement, the tone board in particular uses one plastic peg and the front pots for mounting.
2. Heatsinks are designed like tuning forks. Everytime I worked on the PM7200, I could hear the heatsinks resonating around 1 kHz, and it took a long time to subside, meaning very little damping. There are several fins the same length as well, this can't be good for in terms of vibration/feedback to the electronics. Might try a rubber stip below the heatsink to increase mechancial damping.
3. Generally the case panels are pretty thin (I'm guessing 0.6mm galvanised steel), there's not much rigidity there for resitsing vibration.
4. Use of PCB board connectors everywhere, in particular the tone and volume board are connected using a cheapo board to board connector, this carries the unamplified audio signals.

I could go on but I think you get the idea, on the oustide it does looks resonable, which is probably where most of the reviewer comments come from.

I didn't see any PMs?

It was for Black Octagon.