Hi,
This is a bit of a lost-all-hope post...
i've been trying to diagnose a faulty Marantz AV receiver, SR4300. The problem is that around ~3minutes from switch on, it puts itself into standby. It's always the same time period. Prior to this, it does actually work ok.
It's a pretty good exercise why not to put microprocessors into amplifiers. The standby control is engaged by the micro. I've taken the obvious route and checked its over-current and speaker short protection circuitry, which feeds the microcontroller, however this isn't engaging at all. I'm at a bit of a loss to know what else could trigger the micro into safety shutdown. Marantz say that an over-temperature condition would also trigger it, but I cannot find any thermistors anywhere in the system, and I doubt the micro itself is playing up because everything else seems to be functional.
Would be grateful if anyone else has got experience on this type of amp/problem, or if anyone fancies a good crossword-type problem to solve!
Attached a screenshot of the micro circuit (full service manual here: Marantz SR4300 Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics)
Any suggestions welcome...
Mat
This is a bit of a lost-all-hope post...
i've been trying to diagnose a faulty Marantz AV receiver, SR4300. The problem is that around ~3minutes from switch on, it puts itself into standby. It's always the same time period. Prior to this, it does actually work ok.
It's a pretty good exercise why not to put microprocessors into amplifiers. The standby control is engaged by the micro. I've taken the obvious route and checked its over-current and speaker short protection circuitry, which feeds the microcontroller, however this isn't engaging at all. I'm at a bit of a loss to know what else could trigger the micro into safety shutdown. Marantz say that an over-temperature condition would also trigger it, but I cannot find any thermistors anywhere in the system, and I doubt the micro itself is playing up because everything else seems to be functional.
Would be grateful if anyone else has got experience on this type of amp/problem, or if anyone fancies a good crossword-type problem to solve!
Attached a screenshot of the micro circuit (full service manual here: Marantz SR4300 Service Manual free download,schematics,datasheets,eeprom bins,pcb,repair info for test equipment and electronics)
Any suggestions welcome...
Mat
This kind of fault might drive anybody crazy.
I guess you need to try to isolate the fault. Three minutes sounds to me that something warms up inside, fails, and triggers the standby mode.
You may try out a few things. Either lifting all PCBs connectors that will allow the unit to run ON hopefully no triggering into stdby. Then reconnect one by one until the fault appears again. Turn OFF the unit everytime you re attach a connector in place.
Or you may try cooling down by zone using a freeze spray can for use in electronics. This is specially valuable to track down thermal failures.
I guess you need to try to isolate the fault. Three minutes sounds to me that something warms up inside, fails, and triggers the standby mode.
You may try out a few things. Either lifting all PCBs connectors that will allow the unit to run ON hopefully no triggering into stdby. Then reconnect one by one until the fault appears again. Turn OFF the unit everytime you re attach a connector in place.
Or you may try cooling down by zone using a freeze spray can for use in electronics. This is specially valuable to track down thermal failures.
There is a rumor about these amplifiers and their cousins that after 6-10 years of operation there is failures in the digital board ...Have seen already 4 of them in my shop and most of them was impossible to repair .
If you have a failure there fault finding is not engaged in the classic ways of fault finding ...there is a good chance that you are chasing a ghost there .
Kind regards
Sakis
If you have a failure there fault finding is not engaged in the classic ways of fault finding ...there is a good chance that you are chasing a ghost there .
Kind regards
Sakis
C'mon Sakis, I've seen better concepts from you around in the forum.
We may start out by not mystifying any electronic device including the Marantz gear mentioned.
I repair hi complexity diagnostics medical equipment for a living (ct scanners, etc...) and worked for the big german firm starting with S for about 10 years, and there I learned something valuable, and is that any fault can be tracked down and that electronic ghosts do not exist.
Electronic failures can be traced using whatever valid method to isolate first the stage involved, then the area of the stage to the very component causing the issue.
In the case presented it might involve a very simple troubleshooting to, at least, know the stage causing the trouble.
Besides, the device works fine for a period of minutes until it fails. Whatever normal troubleshooting method may be used: removing boards, lifting connectors, applying a little heat on areas with a hairdryer, freezing components with a freeze can, etc, etc.... He even has the squematics so that is a BIG help.
But if you think you will solve a problem without putting your hands on the hardware then it just won't happen.
Mat, we all would hate to mess with a perfectly running hi tech Marantz, but if it does not work perfectly anymore then you HAVE to put your hands on it, otherwise you will never solve the issue. Just do everything applying your best criterion in other not to cause further issues. For eg, if you decide to apply some heat to a circuit, that means a bit of heat, not melting the components.
If you are not a vast expert on the 4300, just assume any stage may trigger the stdby mode by whatever reason. That's your first premise. Even the RC receiver can do that.
Having said this, your first move might be removing/disconnecting as most stages as it can be done that wouldn't turn the whole device off (isolation method)
Thanks for all your messages.
I should point out that I have been probing this unit extensively for the last 3 evenings. My problem ultimately comes as the microcontroller is performing the shutdown, which I don't have any operational details of. Hence, it's been a case of trying to find whats telling the micro controller to perform the shutdown.
In terms of disconnecting the boards, it simple to disconnect some, but difficult/impossible to isolate down to a single board - they're all quite incestuous. Indeed, the standby/shutdown line eminates from the micro/dsp board, bounces through the main PCB before finally being actioned in the standby PSU board. What I've found by this is:
- video, front panel, radio, preamp boards can all safety be disconnected without freaking the micro out. As for the rest:
- main + micro + amp = shutdown
- main + micro - amp = no shutdown (amp power input disconnected).
This yields an immediate idea that the amp is at fault. However, the feedback loop from the amps power / over current detection simply isn't engaging, so it's a mistery whats going on - other than the PSU must be folding down.
I think I've finally made a breakthrough tonight. The +5VD regulator (NJM7805) is cooking. Liberal doses of freezer spray (thanks for the hint) keeps the system operational. The question now becomes whether it's actually the regular breaking down, or a down-stream component. I've disconnected the main +5VD lead exiting the board and it still shutdowns, but alas this reg also feeds a 2.5 and 3.3 output, which I can't easily pull. Doubt its the 3.3v as its own regulator is ice cold. Best guess from here is to replace the 7805 regulator, then probably a few of its ancillaries. Just take a shot in the dark...
Would appreciate any comments on:
- anyone actually know a 7805 reg to break down?
- suggestions on likely culprits on the downward path (image attached).
thanks again for your support - it's good to bounce ideas around (in addition to the hours spent with scope).
Mat
p.s. ghosts in the system are real where micros are present - it's called software.
I should point out that I have been probing this unit extensively for the last 3 evenings. My problem ultimately comes as the microcontroller is performing the shutdown, which I don't have any operational details of. Hence, it's been a case of trying to find whats telling the micro controller to perform the shutdown.
In terms of disconnecting the boards, it simple to disconnect some, but difficult/impossible to isolate down to a single board - they're all quite incestuous. Indeed, the standby/shutdown line eminates from the micro/dsp board, bounces through the main PCB before finally being actioned in the standby PSU board. What I've found by this is:
- video, front panel, radio, preamp boards can all safety be disconnected without freaking the micro out. As for the rest:
- main + micro + amp = shutdown
- main + micro - amp = no shutdown (amp power input disconnected).
This yields an immediate idea that the amp is at fault. However, the feedback loop from the amps power / over current detection simply isn't engaging, so it's a mistery whats going on - other than the PSU must be folding down.
I think I've finally made a breakthrough tonight. The +5VD regulator (NJM7805) is cooking. Liberal doses of freezer spray (thanks for the hint) keeps the system operational. The question now becomes whether it's actually the regular breaking down, or a down-stream component. I've disconnected the main +5VD lead exiting the board and it still shutdowns, but alas this reg also feeds a 2.5 and 3.3 output, which I can't easily pull. Doubt its the 3.3v as its own regulator is ice cold. Best guess from here is to replace the 7805 regulator, then probably a few of its ancillaries. Just take a shot in the dark...
Would appreciate any comments on:
- anyone actually know a 7805 reg to break down?
- suggestions on likely culprits on the downward path (image attached).
thanks again for your support - it's good to bounce ideas around (in addition to the hours spent with scope).
Mat
p.s. ghosts in the system are real where micros are present - it's called software.
Attachments
Great Mat, looks like you already found the culprit. I don't think the 7805 heats up as being overloaded down stream of some other faulty stage.
The fact that you cool it down and the system does not shut down indicates that the reguator has been working temperature stressed for quite a while and has degraded now.
Just replace it and add a small/bigger heat sink to it to stablish a lower operational temperature.
The fact that you cool it down and the system does not shut down indicates that the reguator has been working temperature stressed for quite a while and has degraded now.
Just replace it and add a small/bigger heat sink to it to stablish a lower operational temperature.
I have seen the NJM versions of the regulators run inexplicably hot, and fail. My Marantz PM-35 amplifier was dead because an NJM7824 regulator was overheating. I replaced it with an OnSemi brand one which runs completely cool.
Incidentally there is nothing wrong with putting microcontrollers in amplifiers, the key is that there should be a diagnostic interface of some kind.. such as RS232, even if it's a header on the motherboard working at TTL level. In this case, the microcontroller's internal brownout circuit will be operating due to low voltage/excessive ripple on it's supply pin, and holding it in reset.
Incidentally there is nothing wrong with putting microcontrollers in amplifiers, the key is that there should be a diagnostic interface of some kind.. such as RS232, even if it's a header on the motherboard working at TTL level. In this case, the microcontroller's internal brownout circuit will be operating due to low voltage/excessive ripple on it's supply pin, and holding it in reset.
Last edited:
Houston: we have lift off!
Finally got the components through (CPC were doing free slow delivery - especially needed for the £1.20 worth I was ordering!). Pulled the main board and did two repairs:
- solder joint had cracked on one of the bulk caps
- new 7805 regulator (used an ST part not NJM)
Bingo, bango, another hour to refit and it's running like an absolute dream!
Thanks to everyone who replied - it's been an interesting repair experience. The funny thing is, had this component not been overheating, I would never have spotted it - the 5v line was still active until standby occurred and it was the microcontroller performing the shutdown, which had no direct connection to this rail. Only possibility is the serial link between the DSP (which was powered from this 5v rail) and the micro went down. I'd never have spotted this without that reg burning. Worth noting if anyone else suffers from what looks like micro shutdown...
Cheers everyone!
Mat
Finally got the components through (CPC were doing free slow delivery - especially needed for the £1.20 worth I was ordering!). Pulled the main board and did two repairs:
- solder joint had cracked on one of the bulk caps
- new 7805 regulator (used an ST part not NJM)
Bingo, bango, another hour to refit and it's running like an absolute dream!
Thanks to everyone who replied - it's been an interesting repair experience. The funny thing is, had this component not been overheating, I would never have spotted it - the 5v line was still active until standby occurred and it was the microcontroller performing the shutdown, which had no direct connection to this rail. Only possibility is the serial link between the DSP (which was powered from this 5v rail) and the micro went down. I'd never have spotted this without that reg burning. Worth noting if anyone else suffers from what looks like micro shutdown...
Cheers everyone!
Mat
It would've been interesting to check the reg output with a scope. Maybe the 100µ 'lytic went dry and caused the regulator to oscillate; a 10n on its own certainly isn't much. Differences in stability between manufacturers wouldn't come unexpected.
This case kinda reminds me of the issues with Onkyo HDMI boards, which more often than not can be revived by replacing some electrolytics near voltage regulators (and sometimes the regs themselves).
This case kinda reminds me of the issues with Onkyo HDMI boards, which more often than not can be revived by replacing some electrolytics near voltage regulators (and sometimes the regs themselves).
Hi, this is a very useful forum. I have the exact same problem with my SR4300. MATM & JAYCEE have provided me with some very useful information. I have had 7805 regulators fail in my Computer Mainframe engineering days. Was in UK, now in OZ, so I think JAYCAR electronics is good here for components. MatM mentioned a dry joint on a bulk cap, I assume either C404 or C405? I suspect any "large mass" electrolytic could crack a joint if the receiver got knocked heavily or bounced when moving house etc, but heat cycling can cause that too I guess. I know this unit can get warm, and folks who put them in enclosed cabinets can cause thermal issues. I will order a couple of LM7805, they hopefully stock the STM brand (although their PDF datasheet is an old 2000 dated National Semiconductor one). Thanks to all, I will report what I find with mine, now I am armed with some very good information. TD.
Update from td47: This was exactly the same issue and probable cause as MATM. The left 15000uF cap had a split/cracked solder joint, so causing lots of instability and ripple, and intermittently dropping into standby. The manufacturer failed to put any silicone adhesive on any of these (but had lots on the smaller unimportant ones!!) , and any shipping would cause this if handled heavily. I did replace both 7805 while I had it aprt, so all should be well (works fine now)).
Feedback/update regarding Marantz SR4300 standby problem
Hi, another update: My 4300 started intermittently dropping to standby again, and would not initialise once brought out of standby (blank display). This turned out to be a dry joint on the RE1 relay on the standby PCB. This is not too difficult to remove, only the top PCB above it needs to come out. I needed a really good light and magnifier to see this cracked solder joint, just discernible when the relay was rocked slightly. It was one of the actuating coil lugs. After re-soldering this and all others to be sure, I put component adhesive on this item, to reduce the risk of joint stress when we move house again!
Hi, another update: My 4300 started intermittently dropping to standby again, and would not initialise once brought out of standby (blank display). This turned out to be a dry joint on the RE1 relay on the standby PCB. This is not too difficult to remove, only the top PCB above it needs to come out. I needed a really good light and magnifier to see this cracked solder joint, just discernible when the relay was rocked slightly. It was one of the actuating coil lugs. After re-soldering this and all others to be sure, I put component adhesive on this item, to reduce the risk of joint stress when we move house again!
Most cracked joints are due to thermal cycling acting on unmatched thermal expansion rates of pcb & components. The lead actually gets yanked up and down through the solder every time the board warms up and cools down, and eventually it crystallizes and cracks around the lead usually. The right strain relief during build is the best preventative.
Thanks for the reply and useful info sbrads. Certainly I see that the electronics PCB manufacturing industry has had big problems with SMT components, particularly with MLCC (Multi Layer Ceramic Capacitors) cracking at the solder pad due to thermal cycling. There is an interesting thesis on this type of thing at:
http://etd.nd.edu/ETD-db/theses/available/etd-09152004-143905/unrestricted/DuffekD092004.pdf
Most of the current research/articles seem to be about SM components. However, older designs, and larger components like large value electrolytics, and relays will tend to be through-hole solder tags. In the case of the SR4300 the standby PCB is at the rear left corner, where there is very little heat generated. Also, the component in question was not "hard down" on the PCB, but had about a half MM clearance, hence I could rock it a bit to see the joint move (so thermal expansion and contraction would not have any vertical "leverage" so to speak). As the damage looked like a lesser version of what I saw in my Sept 1st post (with one of the 15000 UF main smoothing caps loose joint), I still documented it as mechanical handling stress fracture due to no silicone glue restraint - what I had not disclosed is that my stuff was shipped by sea in a container from UK to OZ - and we did have some minor breakages. I think this is the aftermath (what the insurers like to call "Mechanical Derangement")! 🙂 My main aim here is to document the actual item that needed attention for the fault(s) and causes in this thread to help others. As MATTM had the same issue with the large CAPS, it seems that some batches of the 4300 pcb's did not have adhesive applied to the high mass components that really needed it, to reduce the mechanical stresses during shipping.
Yes, there has been loads of research these last 10yrs on SM failure modes, especially with respect to lead-free solder. With through-hole solder failure modes, the NASA research done in the 60's is interesting, they were getting a bit fed up with their rockets' electronics failing it seems, so spent a fortune working out why it happened and how to avoid it.
Best of luck with your Marantz.
Best of luck with your Marantz.
Necrobump!
Just to say thank you to the guys who've worked on this unit - I have one that had an annoying hum issue, I did no diagnosis, ripped straight into it, found the main cap failure, resoldered and rebuilt inside of an hour. Now the hum is about as quiet as the hiss, in the realm of ignorability, I should probably let it warm up and see if it improves!
Just to say thank you to the guys who've worked on this unit - I have one that had an annoying hum issue, I did no diagnosis, ripped straight into it, found the main cap failure, resoldered and rebuilt inside of an hour. Now the hum is about as quiet as the hiss, in the realm of ignorability, I should probably let it warm up and see if it improves!
I've had a few crook 7805's over the years. Usually with varying o/p voltage or noisy o/p's as measured with a CRO. It's nice to see a thread with an actual positive conclusion.
I've had a similar experience with mine going to standby after hours of moderate usage. The 7805's are cool but 7915 & 7815 (-15/+15 regulators) are heating up very fast after 10 minutes. The board around the leads has changed color. I'll try replacing them but am not sure why they are heating up...
- Status
- Not open for further replies.