CR840 silent

My beloved CR840 has gone into protect mode I believe and need a little guidance to repair it. I have owned since new and it is or was in mint condition. I have Tannoy mons in my studio but still like my 840 through my ADS-L620's for casual listening.
I had been listening for about 4 hours with the volume set at about 11 o'clock as I meandered about the house when the amp completely cut out on both channels, cut back in almost immediately, then cut back out to the deafening dreaded silence. Let it cool down and still nothing. Unit powers up, all panel lights and tuner functions appear normal, but the relay making is not happening when energizing the unit. Where do I start and what am I most likely looking to find. A pre driver or power module or some other culprit, and where can I get a good service doc with the working voltages at various points in the circuit to compare as I weave my way through? Am new here and this is my last hope at finding the info I need. I really hope to repair this unit as it has been a steady enjoyable piece of gear. Thanks in advance.

Welcome to AK.
E-mail Norman and tell him what you need. I have always gotten Hi-Q pdf files ASAP.:thmbsp:
http://www.analogalley.com/m_yamaha.htm

Check your Power amp chips voltages. One may be shorted out.

Check out Rob (Merrylander's) thread on the subject. Lots of t-shooting tips.
http://www.audiokarma.org/forums/showpost.php?p=1600326&postcount=1

Hopefully its just a power supply or protect circuit issue.

Welcome to the forum and sorry for your troubles!

I hope you get back to listening to the CR840 soon. Those are really nice units.

Have not had a chance to do voltage checks yet, but thoroughly have gone over the interior and there are no detectable or visible failed components. Does this sound more of a power module or driver module. I will get to voltage checks tonight hopefully. Manual was not listed on Normans page. Did take both supplies from the xfrmr and they were a little bit different from the owners manual schematic, I think, because it is hard to read the schematic. But I looked at the cr saga thread and looks like some good info. Nothing outstanding visually or smell wise, just silence. Well Mackie sub mixer and one of my triamp amps is a ok sub but not as crisp. Have been listening to the system for 30 plus years and appreciate all the help available here from those familiar with the product line.


Say what you mean and mean what you say, then lead or get out of the way.
Me.

Manual was not listed on Normans page.

Click to expand...

E-mail Norman and tell him what you need!! He doesn't have all of his available manuals posted on his website yet !!!!

Just thought I'd offer the opinion that a very powerful trouble-shooting step can be taken by making exactly two voltage readings (one per stereo channel). To wit, the DC voltage at the power amplifier outputs. You won't need a schematic for this - there are test pins provided. Look for the test pins in the neighborhood of the predrivers (the roughly credit-card-sized modules). The measurement should be made with the volume control turned all the way down.

If the DC voltage on either of the test pins is very far from zero, you can be pretty sure that this is causing the protection circuit to keep the speaker output disabled. That is, in this case, the protection circuit is doing its job, and is probably OK itself.

If the voltages are, oh, say, between one tenth (0.1) and one volt, there's at least a faint hope that you can adjust them to zero. The adjustment pots are near the test pins.

If the voltages have greater magnitude - beyond plus or minus one volt or so - suspicion falls on the power amplifier(s).

If the voltages are within the range of plus or minus one tenth (0.1) volt or so, suspicion falls on the protection circuit. (But, you might as well adjust the voltages to as close as possible to zero while you're there.)

Your findings with this test will give you a better idea of what you're facing.

Good luck,

chazix

Test Points.

Ok, I went back through today(8 days a week please) and repeated my visual inspection after realigning the pre drivers (Yamaha seemed very fond of not being too concerned with the vertical alignment of the components) a bit to view what I could not previously see. Also put it on my board assembly table so I could view through the magnifying light and found behind IC702 (right )pre driver a fried resistor. It is R720 which is 390r and looked to be a 1/4 watt, and is tied to pin 11 and also 12 through C716 and 10 through C720 upstream of the resistor and to R722 & 724 on the downstream side, which are the .47r 5w pair on the amp IC704. I have 5% carbon and 1% metal film resistor stocks in 1/4 watt and think I should replace this and ensure the traces under the fried component are still sound, cleaning up the resulting carbon and reconformal coating the board prior to further testing.
Also chazix, very near there are what looks like the same type of physical pin connector towards the front of the amp from R718(2w) 390r with no marking as the TR and TL test points in the protection circuit area. Is it the latter or former you were suggesting I read for the near 0 voltage? It is easier to make a little sense with the bright magnifier. Based on this finding, keep going or new course? I have briefly looked through the continuing saga thread (phew) and think I get it, but will reread more thoroughly and take some notes for reference.

Hi, ppatmandad,

R720's death implies that the power output module failed with an emitter-collector short, which caused one or more of the 0.47 ohm emitter resistors to fail (go open-circuit), which dumped abnormally high voltage across R720. This chain of events will also most likely have damaged the predriver - but perhaps only the section of the predriver that's supposed to limit output current. That is, the voltage amplification function of the predriver might still be OK, in which case replacing the output module and emitter resistors will get you going.

The possible damage to the predriver's current-limiting function might not be all that great a loss. My hunch is that this function never really provided much protection, because it was set up to react too slowly.

The unfortunate thing is that it's tricky to tell whether the predriver suffered additional damage. If you feel ambitious, you'll find a procedure in the "Saga" thread for testing the predriver's main functionality with the power module removed.

I don't have an 840 handy to look at, so I'm not sure about your test pin question. I thought there were only two test pins associated with the power amp section. The ones I had in mind are in the vicinity of the predrivers and DC offset adjustment pots. In any case, it is probably less important to check those test points now. The R720 failure is a clear indicator that an output module has failed.

chazix

Same thing in my old 840...the 390 ohm failed, probably due to high DC voltage from the output amp, the .47 ohm resistor looked failed (burned on the bottom). The output amp may go to a high output voltage if the predriver shorts....the output will try to follow it.

The spkr protect circuit is probably keeping the spkr out relay open because of the high output DC voltage, measuring the previously mentioned TL and TR test points will probably show that. The protect circuit in mine has worked correctly thru 2 predriver failures.

How to fix: obviously replace the 390 ohm, possibly the .47 ohm, more than likely the predriver module, maybe the output amp module......

If you need to buy some modules from the above mentioned sources (in the SAGA thread), you may as well buy the predriver and the output module and do both.

Chazix...any idea why the predriver transistor shorts? I've had 2 of these go on my 840....seems to be a design issue.

I just replaced the entire predriver/output string with a LM3876 amplifier (Texas Instrument part)...not going to keep replacing predrivers, may be a solution for some of you folks.

George

Hi, George,

I guess that, unlike yours, my experiences haven't convinced me that the predrivers have a design flaw. That's not to say that I'd rule it out.

There's a report somewhere in the forum that Sanyo, who produced the Yamaha predrivers, recommended that they be fitted with a heat sink. Some contributors have retro-fitted heat sinks. For myself, I don't think the watt or so that the predriver dissipates should present any problem if the transistors inside it were properly specified.

I don't have any really solid evidence, but my predisposition is to think that the common case is that output module failure leads to predriver failure, not the other way around. In particular, I don't see how any failure mode of the predriver could lead to the output module developing emitter-collector shorts in both output transistors. But that failure in the output module seems to be common, as evidenced by blown emitter resistors.

chazix

Read through "SAGA" thread

I can see no physical .47r emitter resistor damage but will be doing the recommended checks which include removing the output module and installing a couple of 220r resistors from pins 1 & 10 to the common trace of the emitter resistors if I read it correctly. One thing I am curious about though is the schematics I have pulled off the site showing the pre drivers only show 14 values for them not one for each pin. Listed are in order>{.13>0>,13>-39.7>
-40.3>-41.1>2.7>0>0>0>0>-2.99>3.03>41.4}.
The way I am seeing it is there is no value for pin 15. Is this considered to be the +B1 voltage?, which I find no value listed in my nomenclature.
So now, it may be a bit premature but I see the pre drivers are no longer available and there is a schematic and artwork posted that merrylander developed, but also a prototype board actually manufactured into a working part from chazix, but the transistor and resistor count are different than the merrylander. And could not locate on the Barter website parts section. What I gather is the proto shown has a pirated hybrid combining a bit more effective protect circuit morphed into the original Sanyo design. Info? And if I have not expressed so, seriously, thank you gentleman for your generous sharing of your technical saavy and information. There are also a couple of electrolytics located in the power supply area I am leary of due to it appearing the clear coating (never seen before on to flat of caps) appears to have cracks opened up on the seams of the couple of the triangles the flattop rupture caps manufactured into them. They are the largest of the small electrolytics and will try to id on schemo. I migtht just pull them and put them on the cap meter. But first the amp pull and fake out resistors to check the pre driver and protect circuit.
Oh, dim bulb setup? what and why. Lower line voltage? Can I use variac instead?
But first off a horizontal adjustment behind the eyelids as it has been a long long week, and its' only Tuesday. 40 winks won't due.

Hi, ppatmandad, sounds to me like you've done a heck of a job absorbing all the arcanery in the saga - I'd need to sleep after that, too!

Your 0.47 ohmers may not look damaged, but it's virtually certain that at least one is. The toasted 390 ohm is, pardon the expression, a smoking gun that points to one of the 0.47s having opened up. Best to replace the 0.47s in pairs, because the excessive current will have gone through both.

The reason there are only 14 figures in your list of predriver voltage readings is that pin 5 is unused.

The dim bulb jazz refers to inserting an ordinary light bulb in series with the AC supply to the unit under test. Although it will reduce line voltage some, in my mind the key thing is that it both limits AC current, and gives you a visual indication of whether a lot of current is being drawn. I think a dim bulb rig is preferable to a variac in general for solid state gear (unless maybe the variac has a current meter). For a CR-840, I'd go with a bulb between 60 and 100W.

The test with the tacked-in resistors should tell you if your predriver is predriving. If it's not, and you decide to replace it, you can PM me if you're interested in one of my repro modules. Or, you can see if you can find an original IG02940 from a supplier you're comfortable with. Offerings occasionally show up on that auction site. I wouldn't recommend substituting an IG02950 (CR-640 predriver) - it would function, but presumably isn't designed for the higher voltages found in the CR-840.

Good luck!

chazix

Thanx chazix, yep, could be the output module going first, I don't remember, when I fixed it the first time it was almost 30 years ago (wow). It then died a while later, I don't remember when that was....

In any case, as mentioned the 390 ohm and at least one .47 ohm were damaged I knew the amplifier string needed repair but was not about to find the parts, etc. only to suffer another failure, hence I designed in the TI (National) LM3876 and eliminated all those other parts in the process.

It works, probably won't suffer the same fate, plus I have a bunch of them.

George

As suggested .

So, had a little spare time and pulled R720 and checked and cleaned the traces on the board and all looked ok. Checked emitter resistors and found the forward one open. Bummer. So I pulled out the amplifier module and cleaned up the bottom of the board. Anyone ever notice that some of the installation quality on their Yamaha leaves a little bit to be desired as far as neatness and attention to quality in the form of dried crusty flux not cleaned after soldering was complete. Where I come from this is something that would be frowned upon. Anyway, before I pulled the amp module I took some cold resistance readings to compare against the left channel and there is definitely differences between the two. I haven't had a chance to solder in the simulation resistors as a workaround for the removed amp but suspect I will see the differences in the predriver cold resistances are leading to a predriver retrofit. Now, if I find as suspected, how do we work out a predriver board or setup. Chazix, you have been the voice of experience on this and what can you tell me about your protection mod for the stock predriver preamp circuit. And how do we connect? Also afterthought edit, do the emitter resistors stay installed for the 220r test?

Hi, ppatmandad,

For the "simulation" test, it doesn't matter whether the emitter resistors are present or not - they are effectively out of the picture anyway, when the power module is removed.

If you find you need a predriver, and want to look into using one of my repros, contact me by clicking on my screen name above left, and pick "Send a private message". (It's kinda like email, except we don't need to expose our email addresses here where spam-bots might feast upon them.)

Here is too much information about the part of the predriver that acts to limit current in the output module: I copied a very similar chunk of circuitry from the Marantz SR-6000 to use in my predriver repro modules. This decision was based as much on superstition as engineering, to be honest. The Marantz power amplifier section is very similar to a Yamaha CR-1040's, but the Marantz is not notorious (as far as I know) for frequent failures. The only thing that's obvious to me about the differences between the two current-limiting circuits is that the Marantz version uses a lot smaller capacitor to set the response time - that is, the Marantz limiting responds faster than the Yamaha. For my repro modules, I picked a cap value that's in between, because with the Marantz-sized cap, the Yamaha would "ring" a bit when going into limiting. But the value I settled on is way closer to the Marantz than to the Yamaha.

Cheers,

chazix

Finally had time

I had a spare moment last evening, finally, and saddled up to the test bench. I replaced R720 and installed a pair of 220r resistors from the emitter resistor common to pins 1 and 10 on the bad channel. I had previously pulled the amp off that channel and rang out pins 2/3 and 8/9 and both were 1 ohm or less.
So, R720 got replaced and added the fake em out resistors and plugged it in, fully intending to plug in the variac and put a clamp on in the circuit but it was late and forgot. So I hit the power button, so far so good. The magic smoke didn't escape. And it seemed like a long time but it was the normal delay and the protection relay clicked in. Cycled it on and off several times and relay appears to function properly. So, the question is this. Can I take the rest of the pre-driver pin readings in this configuration to determine if the pre driver is indeed good or do I need to get my replacement emitter resistors (1 open) and amp and restore the circuit to test the pre driver. To say I was surprised when the relay cycled is an understatement. Take readings as is or restore and take them?
Also, looking at the amp specs it seems almost that the 840 uses a module between the 060II and the 070II. Which one should I order? And have not found similar resistors. Any direction or vendor choice?

Hi, ppatmandad, nice going!

You can go ahead and make tests without replacing anything else. What you want to confirm is that the center point of the two fake-em-out resistors is at pretty close to zero volts, and that the voltage across each of those resistors is about 1.3 volts. That's an approximate figure - what's more important is that the two resistor voltage readings should match very closely.

If the DC voltage checks come out OK, it would be a good idea to try a HEADPHONE listening test. (The fake-em-out arrangement won't be able to deal with the load presented by speakers.) Listen to a trusted source at low levels, and make sure both channels are undistorted and free of noise.

If all these checks come out OK, you can be very confident that the main function of the predriver (voltage amplification) is OK.

The predriver's secondary function (output stage current limiting) might not be working, but my belief is that this function was never very effective anyway, so it's no loss. Note that this function is aimed at saving the amplifier from external problems such as shorted speaker wires. The other aspect of protection - saving your speakers from an internal amplifier problem - is handled elsewhere, by the circuitry that contols the relay.

For the output module replacement, best to go up to STK0070II. (Or higher, but in my opinion there's no added safety value.) Get it from a trusted supplier. I've been happy with B&D. Whatever supplier you use will most likely also be able to source the emitter resistors, too.

chazix

Powered up and no smoke

First, Chazix, thank you for your assistance and willingness to share your expertise. Ok, was not sure where to measure for the center to? but I did set the meter to mv dc and hooked up to the TR and spkr (-). Started out in the
-200 or so mv and dropped down to just over -30 mv under no load with the volume fully down. It took a while to drop and then just hovered :thmbsp: Next took a drop across both temporary resistors and they were almost exactly the same at 1.34_v apiece. For the zero measurement would that also be against the speaker (-)?. Have to run over to folks house real quick but will do a headphone test upon return this evening and will edit this post upon execution. So far so good.
Question, is the enough excess capacity in the power supply somewhere to be able to put in a small fan blowing across the predrivers towards the amps?
Can get fans almost any voltage these days and I have fans either in the gear or mounted right near them in all my live performance equipment. I do this because of a daytime outdoor gig many years ago with the sun cooking my Hiwatt head to the point that solder melted. But even though not tubes it seems that the engineers modeled this one a little close to the live die line to save a few pennies, and a fan may benefit the big chips.

Based on suggestions by Merrylander, I put a couple heatsinks on the pre-drivers. picture is in the 840 saga thread. The passive cooling should be plenty to preserve their life.