Marantz 2270 Slow to Click in...sometimes

Thank you petehall347 and NAD80 about the information regarding the flyback diode, I will crack it open again soon and check that out and also take some photos of the PS caps, though I'm not sure the difference between them will show up, it's very subtle.

Also, thanks Robisme, yes I checked the DC offset and while neither was far off, I zeroed them out while I was in there, but still get that occasional scratchy sound. The issue is so intermittent however, I am not sure that I could definitely know if a test with the preamp separated from the amplifier would be definitive, but I will definitely try, in case I get a positive result.
 
The buzz you get right before the relay opens finally is the relay buzzing due to a low voltage condition, caused by the Cap/Transistor combination that turns it on and off. The transistor is used as a switch to open and close the relay. It's supposed ot snap open and closed. In your case the voltages are coming in low, ramping up and causing the relay to vibrate rather than engage completely until the voltage is high enough to cause the electromagnet to close the contacts. Make sure the supply voltage up to the relay is correct, and if not correct it as needed. If the cap in the circuit is leaky, it's probably causing the ramp up. Make sure the NEW CAP is oriented correctly, and replace the transistor with a modern replacement.
 
Agree the relay should not buzz even if operated slowly by the circuit. If it buzzes then the +24V should be faulty. High ripple most likely. One of the 2 rectifier diodes may be bad (H811) or C807 could also be bad.
 
Thanks for all the assistance, I sure do appreciate it!

The ramp up buzzing was there before and after I recapped the board, so my guess is that is isn't C807.
I am going to open it up again next weekend and check the H813 as instructed, take photos of the PS caps as requested, and depending on what comes of it, possibly replace H808, H809, & H810 as suggested. I know for a fact I don't have all the parts I'll need so I think I am just going to order parts for the whole board while I'm at it, just so I have them on hand if I need them. I don't plan on leaving the 2270 on the bench opened up any longer than I have to, since it sounds so awesome! :)

Based on my research, here is the replacement parts list I came up with, does everything look copacetic to y'all?

H801: Diode SIB01-02 --> UF4004
H802: Diode SIB01-02 --> UF4004
H804: Diode BZ-140(14V 1W) --> 1N5351BRLG
H805: Transistor 2SC696 --> BD241CTU
H806: Transistor 2SC734 --> KSC1845
H807: Diode WZ-071 (0.5W 7.1V +0.4) --> BZX79-C6V8 T/R
H808: Transistor 2SC945Q --> KSC945CGT
H809: Transistor 2SC1213A --> KSC2383YTA
H810: Transistor 2SC1213A --> KSC2383YTA
H811: Diode SIB01-02 --> UF4004
H812: Diode SIB01-02 --> UF4004
H813: Diode SIB01-02 --> UF4004

I'm trying to stick with Mouser if possible, since I know for sure that everything will arrive before I need them (Mouser ships from pretty close to where I live).
 
It doesn't really sound like a faulty relay, but for the few extra $'s I'd probably add a new relay to the list as well.
 
The question regarding this repair but applicable to any other as well, is do you want to reapir the thing, or do you want to exchange parts hoping the issue gets solved without knowing what was really the root cause?

Grailseeker will tell us what voltage value is present at the relay (relative to ground), either activated or deactivated. It does not make a difference.

Second interesting thing would be to isolate the speaker protection circuit from the amplifier, so it does not interact. That way would be easier to find the problem. Anyways the circuit is quite simple. If the supply voltage is OK, then a couple readings on the transistors would easily determine where's the issue. No need to change everything.
 
I'll chime in here and mention that I've had a very similar issue with a 2270 before. H809 and H810 had very low gains compared to what they should be, and replacing them each with KSC2383 remedied my problem.

I'm currently restoring one right now and was looking for a good sub for H806. I like to replace transistors when I do restorations. In my time repairing stuff, I've found bad transistors far more often than I've found bad capacitors. My experience is a bit limited, though.
 
The question regarding this repair but applicable to any other as well, is do you want to reapir the thing, or do you want to exchange parts hoping the issue gets solved without knowing what was really the root cause?

This is a really good question and I've been thinking a lot about it since you posted it. Generally speaking I am of the disposition that if I'm going to do something, I should do it right. So, in that respect, I should think it would be best to isolate the problem and solve without replacing everything.

That said, I generally operate under the premise that whenever I work on anything, there's always the possibility that I will introduce an error or exacerbate the problem, so while I take every precaution, I have to concede there's a chance things might go south.

With that in mind, I also wonder: if a part has failed due to age or known unreliability, wouldn't it behoove me to switch out other nearby components to help mitigate future problems? Or should I stick by the age-old adage, "if it ain't broke, don't fix it?" Isn't recapping the whole board already somewhat committing to pre-empting part failure?

All of that is to say that I'm of two minds. Ultimately, I just want a reliable, working 2270, however I can make it happen. But also yes, I want to only fix what needs fixing. But maybe also trying to reduce the likelihood that I'll have to go back in again later. I'm really not sure the right balance.
 
Really, you should stick to replace the faulty item, and not a entire list of them, in hope of avoiding an occasional failure. Now in 2017, I'm turning 25 years as FSE servicing medical equipment, mainly CT and MR scanners. I'm still servicing 20 y/o stuff, some even older, and you never ever see a electrolytic cap go bad. The worst thing for these caps is to not be energized for a long period of time. Otherwise they'd run for ever.

Where you can expect cap failure due to age is with tantalum type ones. Going for a full recap thinking your gear will improve dramatically is just plain fantasy. Unless a cap is really faulty then you should expect nothing else but -hopefully- the same performance. The difference between 2 devices, one of them noticeably better than the other, is mainly in topology, and/or the power supply.

As far as my experience tells, failure due to known reliability is due to poor SOA chosen for that given component. In that case you should just improve whatever was wrong in the first place, by a more suitable component. Ever saw a resistor burning itself and the board beneath through time, slowly? That's a good example of a improper component for the task assigned. May be the rest is just perfect and would last a lifetime.
 
OK, got my supply of parts and back at it.
Below are two images, one indicating voltages at the relay. Looks like the supply voltage is 28V (it stabilizes down from about 30V on startup)
The other image attempts to show the ever-so-slight physical convex top on one of the filter caps that was mentioned earlier.

2prheom.jpg


Do these voltages seem correct?

2qbrn10.jpg

I'm really not sure I captured the difference, but the one on the right is the one in question.
 
And here's something else I am noticing. The idle current doesn't really stabilize, it creeps up, very slowly, but definitely creeping.
 
The idle current, you mean the bias?

I would not worry about that capacitor above. Yet, if your MM has capacitor measurement capability, remove and test it. Otherwise forget it. If that cap was bad the amp would hum like mad.

I don't think that board above where you embeded voltage values is in the service manual I have. Can you move them onto the diagram below?

Sin título.jpg
 
Yes, I was referring to the bias. It starts at about 5mV when turned all the way down according to the measurement procedure in the service manual, but after several seconds it ticks up a little, and then more, and more until it surpasses the 10mV value (I have a SN above 3900 so I am measuring J754 & J760, aiming for 10mV) and can no longer be lowered into range. I am not sure the upper limit or if it ever stabilizes. I will try to get additional information about that today.

More about that in a minute, first, here are those initial measurements in the snippet of the schematic you sent.
29dzgp4.jpg


However, I do have an update. I'm pretty sure I have figured out the relay issue, at least on the surface. It seems H809 was the culprit. Based on research from aidanhunter's comment above (#29), I swapped out the old 2SC1213A for a KSC2383YTA and the relay now snaps in cleanly after about 2 seconds with no more ramping up. I also switched out H810 to keep things matched. Additionally the voltage measured at the relay is now 26.7V instead of 28V

Honestly, if it weren't for the rising bias value, I would think this was a grand success. I didn't notice that value until I was doing the final adjustments, about to close it up and revel in my triumph, so I wasn't sure if I was the one that introduced the problem. To find out, I then swapped back in the 1213s and checked again. It seems that other than having a lower initial floor, the value rises anyway. So, I put the KSC2383s back in, the relay is great, the sound is (as usual) great, but my guess is that this rising current has been there since I got it.

Could this be a reason for the old caps on the P800 being leaky and cooked onto the board?

I'm starting to think there was no "tech" that looked at this before it was sold to me.

Does anyone have any insight into why the bias is behaving like that? How much of a problem is it?
 
UPDATE: After about two hours, the bias seems to be sort of stable around 15mV. It started at a bout 4mV, and crept up to about 12mV in the first half hour and then crept up slowly for the next hour, peaking at about 16mV. The it sank back down to around 15mV and has been in that vicinity ever since. This is with the trim pot turned all the way down.
 
Bias creeps around. That's why service manuals say for you to set it after the amp has been turned on for awhile. 2270 bias is particularly flaky though. There's a good thread on this topic here. Read this and try to understand it: http://www.audiokarma.org/forums/index.php?threads/marantz-2270-bias-wont-go-below-15mv.322206/

So, based on this, I add 5.6 ohm 1W resistors in place of those wires when I work on a 2270. It makes the bias much more stable and is far less complicated than relocating that diode to one of the driver transistor heatsinks.
 
Thank you all very much for all the help with the speaker protection issue! It is very pleasant to hear it just click in after only a couple seconds! I am grateful for all the shared expertise!

aidanhunter: Regarding the bias issue, thank you for directing me to that article, which sounds like it has become a classic on AK. I have read it twice now and, if I'm being honest, I am barely able to keep up on understanding it at this time. I have been doing a lot of research to try to get a better grapple on it, so we'll see how that goes. In the meantime, it appears I am having much the same issue as the OP from that post in that I can't get it to bias below 15mV. Once it stabilizes (at least 45 minutes apparently) it, it doesn't seem to drift much more than .5mV in either direction. Sounds like the solution in that case is maybe an unavailable part? (Motorola MR500)

With that in mind, I wonder would the 5.6K 1W resistor replacements you suggested help in this regard or would you say there are other things happening?

I supposed my final question generally is if this amp's bias is 5mV over the spec (when turned full ccw), how much of a problem am I looking at? I know some folks run stuff a little hot intentionally.
 
Back
Top Bottom