(Another) Troubleshooting a Yamaha receiver with DC_PRT (protect shutdown) probl

[NHcoder]

PS I hope I don't sound too abrupt. I am an Engineer (retired) i.e. "on the spectrum".

Absolutely no complaints here.

[alanep]

We're talking about checking continuity/resistance between chassis ground and the "E" marked pins on those two connectors, correct?

(learning how to quote, I don't know how to the get the originators name in there)

Yes please, if you can measure the resistance between the "E" pin on 4W1041,2 W1042, W1043 and chassis ground.

[alanep]

Please note incorrect W1041. Should be W1043. Fixed in original posting.

[NHcoder]

We're talking about checking continuity/resistance between chassis ground and the "E" marked pins on those two connectors, correct?

(learning how to quote, I don't know how to the get the originators name in there)

phpBB is a little esoteric, but I've been posting on forums like this for many years, happy to offer what little help I can in the other direction.  If you click the quote button on the message you want to quote, it will quote the whole thing, and the name will show up as well as the clickable link to the quoted post.  You can copy and paste that line over and over to quote different sections, just make sure you close each one as well with

Code: [Select]

[/quote]
You can also type out the name or copy/paste it in a simple tag like this:

Code: [Select]

[quote author=alanep]
Yes please, if you can measure the resistance between the "E" pin on W1041,42 and chassis ground.
[/quote]

Which makes this:

Yes please, if you can measure the resistance between the "E" pin on W1041,42 and chassis ground.

I was able to test W1043 to both pins since that board sticks out past the heatsink, and it's clean, 0 ohms to the chassis.  W1042 (Main(2), goes to CB707 on Function(1), page 114) is 0 ohms as well, but I can't get access to the pins on that one, just the test jumper.  Just to be sure on W1042 I checked from the jumper to both pins at the other end of the connector and it was 0.0 on both of them as well.

Note I checked W1043, not W1041, maybe that was just a typo?  1041 is a bigger connector with the DC_PRT line and other PRTs, an SPI connection, etc.

Please note incorrect W1041. Should be W1043. Fixed in original posting.

Yep, caught that.

[alanep]

Yes, it's W1043 (fixed previous postings)

Finding W1042, W1043 "E" pins measuring 0 ohms to chassis is absolutely bizarre i.e. you'd swear that would have been the problem.

We came to same conclusion/measurement previously (I'll have to look up the post). That was one of the reasons I thought of the unity gain voltage measurements/sanity check (which you may still wish to do).

Thanks for info re quoting.

[alanep]

One more thing just out of left base (and one of the most fundamental things to check first). I'm not sure if the power supply rails were tested (maybe a big oops)....

Could you check/measure +56V, -56V on the power amplifiers somewhere that's convenient.

[alanep]

Continuing on with power amplifier power supplies, there are actually 4 feeds to each power amplifier. There's +/-56V unregulated to the output transistors and also +55.6V and -53.8V (from regulators) to the driver stages.

I've got a hunch, it could be one of the regulators...

Updated below:

I've added the schematics (all extracts from page 138).

A low voltage on the driver part of the amplifier could certainly cause these effects. Just as an aside, it would have shown up on unity gain voltage measurements...

[NHcoder]

Continuing on with power amplifier power supplies, there are actually 4 feeds to each power amplifier. There's +/-56V unregulated to the outputs and also a +55.6V and -53.8V (from regulators) to the driver stages.

I've got a hunch, it could be one of regulators...

Ok.  I've been searching for test points I can access.  I also am pretty sure I can get access to test those resistors directly, I didn't even notice the vertical jumpers you mentioned before, maybe I was too tired.  I believe I can get at them all though as they extend up to where the big inductors are, and I can access that area pretty easily.

[alanep]

I've updated the previous post to include the schematics.

If you want a break for the night (not sure what time it is there, 3:30pm or so here). I'll be away tomorrow (bushwalking).

I think there's something to go on with.

[NHcoder]

I've updated the previous post to include the schematics.

If you want a break for the night (not sure what time it is there, 3:30pm or so here). I'll be away tomorrow (bushwalking).

I think there's something to go on with.

It's almost 1am here but it's Friday night, so I have time to spare.  Sorry for leaving you hanging, getting access to those pins isn't easy, but I think you're on to something.  I was able to trace the paths to test point jumpers for some of them, and all of them I could get to are showing bad values.  I can get to the rest but I'll need to power the receiver off, unplug it, and attach some of my probe clips with it off so I don't risk shorting something trying to maneuver them in.  It's really a no go on all but one of them with the normal meter probe.

Anyway, the result:

W1001 on MAIN(1) only gave me access to -LB_L @ J1333, and I measured -2.47v.

W1003 on MAIN(2) gave me access to both +LB_R @ J1298 and -LB_R @ J1211.  These were -0.6v and -2.5v respectively.

This is in the default diagnostic mode, in which there is an output DC voltage present at the speaker terminals, in case that matters.  I haven't dug in yet to see what's supplying these rails but I imagine it's time to tear the boards back out and test whatever it is directly.  I wanted to get back with an update before proceeding just so you didn't think I'd disappeared.

[alanep]

Ok, it's up to you but mistakes get made when you're tired...

With those voltages (I'll try quoting eventually), there's definitely something wrong with the output of the regulator.

I'll have to look at the layout as to where the regulator actually is (updated: page 116).

The regulator is a discrete design but there are a couple of things:

1. The negative regulator mirrors (almost) whatever the positive regulator is outputting.
2. The positive regulator output voltage can be changed by a signal "AMP_LMT"

Updated:

If the "AMP_LMT" signal goes to 0V, it will turn off Q1096 & subsequently the series pass Q1009 i.e. shutting down the regulator.

Updated:

So if possible, the "AMP_LMT" voltage should be measured. It's on W1041 pin 3. Component overlay - Main(3) (page 116 with page co-ordinates C6)

Updated:

If it's any easier, "AMP_LMT" is also available on IC908 pin 4 or connector CB909 pin 3 (page 132).

[NHcoder]

Ok, it's up to you but mistakes get made when you're tired...

With those voltages (I'll try quoting eventually), there's definitely something wrong with the output of the regulator.

I'll have to look at the layout as where the regulator actually is (updated: page 116).

The regulator is a discrete design but there are a couple of things:

1. The negative regulator mirrors (almost) whatever the positive regulator is outputting.
2. The positive regulator output voltage can be changed by a signal "AMP_LMT"

Updated:

If the "AMP_LMT" signal goes to 0V, it will turn off Q1096 & subsequently the series pass Q1009 i.e. shutting down the regulator.

Updated:

So if possible, the "AMP_LMT" voltage should be measured. It's on W1041 pin 3. Component overlay - Main(3) (page 116 with page co-ordinates C6)

Hmmm.  W1041 isn't something I have access to since MAIN(3) is the guy buried at the bottom of the stack, but fortunately that cable just runs up to DIGITAL(1), the very top board, which I can access easily @ CB909, which shows 1.8v on AMP_LMT.  The manual says that's a 2N5551C-AT and the datasheet for that says it's in saturation at 1v, which I would say points to something wrong on the MAIN(3) board itself.

On the other hand, the schematic says the base of Q1096 should be 3.1v, not 1.8v.

EDIT: Probably sleep time for me soon, but again, thanks again for all the help. 

[alanep]

Yes, I did update the previous post to include CB909. What is the etiquette when a post is modified i.e. do you add a new post saying the previous was modified or should the new post quote the previous post?
EDIT: Ok I get the gist (use EDIT).

1.8V is not good. It comes straight out of IC908 pin 4 (the main controller IC page 132) i.e. the IC is faulty, something is telling it to power limit or:

There is one other small possibility that pin 4 is being driven low (unintentionally) by a fault with Q1096. There's still something strange, even with 1.8V (in lieu of 3.1V), the positive regulator should keep same the ratio i.e. 3.1V reference produces 56V, then 1.8V should produce 37.8V (which it's not doing). A short explanation: whatever voltage is on Q1096 base should also be on Q1097 base (differential pair). The voltage on Q1097 base is determined by the ratio of R1283 and R1284 (in voltage attenuator mode 10K/(200K+10K) = 0.0476). Another way of looking at it is the regulator output should be approx. 21 times the voltage at the Q1096 base.

Anyhow, time for a bit more of a think about it & for you, to get some sleep.

[alanep]

I'll be away for the next 24 hours.

Another thought (for another day): if Q1096 had an open circuit collector, Q1009 wouldn't turn on (positive regulator output voltage would be near 0V) and more importantly, it would load down IC908 pin 4 (AMP_LMT) with R1282 (560ohm) less the Q1096 Vbe (which is a lot). This could explain the lower 1.8V.

Q1096 is probably in the bowels of the amplifier but it might be worth removing and testing. Do you have anything to check transistors? You can use a multimeter (digital on diode mode, analogue on x10 resistance range) to check the base to emitter & the base to collector junctions (they behave like diodes in these tests). I'm hoping the base to collector junction is open circuit. Note, the x10 resistance range on an analogue multimeter is to keep the current down - the x1 resistance range on a 20Kohm/volt meter passes something like 150mA full scale which may blow the junction.

I did a quick calculation on Q1096's power dissipation and it's about 140mW i.e. still within its 625mW rating but it's getting up there as far as dissipation in free air. Maybe, just maybe Q1096 might be faulty...

EDIT:
Possibly another way of testing Q1096 without having to remove it. The idea is to deliberately feed it with 3V. You'll need to disconnect the IC908 pin 4 (ARM_LMT) wire from CB909 & use the wire to feed 3V (maybe 2 x 1.5V batteries) down to Q1096's base. The negative end of the 3V source should be connected to the chassis. If the regulator doesn't turn on then Q1096 is most likely faulty (or something else is faulty in the regulator) & you will then have to go into the bowels of chassis to fix it.

Also if the regulator does turn on, keep it as brief as possible as it might have an actual over current problem (when the driver stage powers up).

[shakalnokturn]

If you go to the trouble of disassembling also check R1008, R1009 (they are fuses), R1283, R1285, R1288 just because high value resistors tend to go open with such voltages.

[alanep]

If you go to the trouble of disassembling also check R1008, R1009 (they are fuses), R1283, R1285, R1288 just because high value resistors tend to go open with such voltages.

Really good point shakalnokturn, with R1009 open, there won't be any voltage on Q1096's collector which will cause loading of the ARM_LMT line (similar to a blown collector junction).

If R1009 was open then finally, the following would also make sense (no ratioing):

There's still something strange, even with 1.8V (in lieu of 3.1V), the positive regulator should keep same the ratio i.e. 3.1V reference produces 56V, then 1.8V should produce 37.8V (which it's not doing).

So maybe bite the bullet, skip the testing of Q1096 and go straight to R1009 - all the dots do seem to be lined up (& while you're in there, check R1008, R1283, R1285, R1288 as per shakalnokturn's suggestion).

EDIT:
If R1009 is blown, it begs the next question: why did it blow? Maybe thermal cycling or is there another fault?

R1008, R1009 are 10ohm 1/4 watt fuse resistors i.e. 158mA through them (dropping 1.58V) will hit the 1/4 watt dissipation. Wait & see?

[NHcoder]

Ok guys, thanks.  I'm going to just tear it apart again and check the resistors & transistors on MAIN(3) and will let you know how it all goes.  More detailed reply follows.

Q1096 is probably in the bowels of the amplifier but it might be worth removing and testing. Do you have anything to check transistors?

Yeah, it is, but I do a lot of hobbyist digital work so I should be able to test the transistor safely in circuit.

Possibly another way of testing Q1096 without having to remove it. The idea is to deliberately feed it with 3V. You'll need to disconnect the IC908 pin 4 (ARM_LMT) wire from CB909 & use the wire to feed 3V (maybe 2 x 1.5V batteries) down to Q1096's base

I thought of this too but the risks seem a bit too high for now, and CB909 is in a package that's at the very edge of my ability to reliably solder individual pins on.

If you go to the trouble of disassembling also check R1008, R1009 (they are fuses), R1283, R1285, R1288 just because high value resistors tend to go open with such voltages.

Will do.  I'll check everything I (or you two) can think of while it's apart before putting it back together.  If that means checking every discrete component, so be it.

So maybe bite the bullet, skip the testing of Q1096 and go straight to R1009 - all the dots do seem to be lined up (& while you're in there, check R1008, R1283, R1285, R1288 as per shakalnokturn's suggestion).

It's all on the same board.  Once it's out, no reason to not test all of it.   Enjoy your day out in the sun, if you can find any!

[NHcoder]

Small update.  Went through some of the testing mentioned already and results are as follows.

• R008 and R009 both seem fine.  Good connections, no discoloration or anything like that.  R1008 reads 10 ohms, R1009 9.9 ohms.
• R1283, R1285, and R1288 are difficult to test in circuit but seem like they're probably ok as well.  The results I got are consistent with an "ok" resistor in parallel with a bunch of other crap, in my experience.  R1283 was very spotty read but seemed to be around 100k, stated value is 200k.  R1285 read 65k, stated value 100k, and R1288 read 118k, stated value 100k.
• Q1096 seems fine as well.  At first this one confused the hell out of me but it turns out it (2n5551c) just has a weird pinout. where the center pin is the collector rather than the gate.  I got 0.7v drops on both base-emitter and base-collector, and nothing on all other combinations.  This satisfied me personally but if you think I need actual functionality, I can do so.

The receiver is in pieces now so can't be tested in a powered on state, but everything is more or less accessible for offline testing.  I'm going to take a break from this for a bit and come back to it later this afternoon or evening.

Once again I appreciate all the assistance. 

[alanep]

This is a slippery one. From my point of view, it is behaving like the regulator is not receiving power hence the focus on R1009. Can you continuity check that one side of R1009 is connected to B+ (same positive supply as output transistors)?

Not sure if B+ was actually tested i.e. the main +56V rail.

I'm away now for 10 hours or so (hiking/bushwalking).

[NHcoder]

This is a slippery one. From my point of view, it is behaving like the regulator is not receiving power hence the focus on R1009. Can you continuity check that one side of R1009 is connected to B+ (same positive supply as output transistors)?

Not sure if B+ was actually tested i.e. the main +56V rail.

I'm away now for 10 hours or so (hiking/bushwalking).

I believe the B+ connection is ok, I tested continuity between R1009 to J1001 which is nearby, and then to a pin on another test port or something on MAIN(3) called CB109.  It's clear on the other side of the board, and is the path R1009 takes to one of the big caps, C1105. 0.0 ohms all the way.

[NHcoder]

I messed around with this a bit more today and with it all apart I thought I'd come at it from another angle, checking for good power at every stage starting from the main board where the mains comes in.  We have been having what I would call "bad luck" trying to trace backwards from the outputs to their supply so I thought it might be worthwhile to try the other direction.  I know that this board (POWER(1)) must be at least mostly good because it runs basically all the digital stuff including the display and power button, but I wanted to check the output of the power transformer since I have access to it now.  This hasn't gone as planned unfortunately so I think I'm missing something on that schematic.

When mains is present, the 5.5v output on POWER(1) comes alive and feeds a currently disconnected output port (W2001).  I used this 5.5v supply to feed a 3.3v linear regulator on a breadboard, and fed the 3.3v back in on the neighboring connector W2505, to the +3.3M, ACPWR_DET, and MCPU_N_RST pins where it should be present if the other control boards were hooked up.  I then tried activating the transistor Q2004 with 3.3v through a 1n4001 on the PRY pin.  This pin says it takes 2.7, which itself comes from a diode at the 3.3v output of an IC on another board.  I figure the 1n4001 would be ok as it has a drop of about 1.1v, but maybe 2.2 isn't quite high enough.  I was hoping to hear an audible click from the relay that Q2004 is controlling, which activates the main power transformer, but got nothing.

Going to get some sleep now, maybe try a diode with a smaller drop tomorrow, or try the 3.3v direct.  I believe the datasheet for Q2004 said up to 5v on the gate is ok, but I'll have to double check.  Attached an image of the power board schematic I'm working from.

[cantata.tech]

I had the same problem with my Yamaha receiver.

Unfortunately my patience was less and I decided to break the problem into two logical parts.



Obviously there is more to this story than just being clumsy. I just gave up on Yamaha amplifiers and switched to a different brand.

I will use the Amplifer and Power section somewhere else.

[alanep]

I'm back from the bushwalk so I'm a bit stuffed but I did think of a few things.

The previous post is one solution to frustration (with a Class D O/P stage?), I hope we can find a less brutal option...

Firstly, in relation to turning on the power transformer:

I then tried activating the transistor Q2004 with 3.3v through a 1n4001 on the PRY pin.  This pin says it takes 2.7, which itself comes from a diode at the 3.3v output of an IC on another board.  I figure the 1n4001 would be ok as it has a drop of about 1.1v, but maybe 2.2 isn't quite high enough.  I was hoping to hear an audible click from the relay that Q2004 is controlling, which activates the main power transformer, but got nothing.

You could put 3.3V or more (see below) on the Power (1) PRY pin (W2505) as long as the cable is disconnected from the connector (so it can't back feed to whatever is driving it). Q2004 base current is limited by R2050 (page 139) i.e. it's not critical although it's best not to go lower than the original 2.7V or higher than say 5.5V. The Power (1) board has got mains potential so please be careful.

To check that RL201 is operating, I can think of a couple of options:

1. You could measure Q2004's collector voltage which would also indirectly measure the relay coil voltage.
With PRY 0V, the Q2004 will be off & its collector voltage should be 5.5V (pulled up by the relay coil) & since both RL201's coil connections are at 5.5V, it will be de-energised (no current flowing through the coil). With PRY 3.3V or higher, it's collector voltage should drop to 0V i.e. one connection of RL201's coil will be at 5.5V and the other connection (Q2004's collector) will be 0V i.e. in theory, the relay will be energised.

2. Find a reasonably safe low voltage secondary and check it with a multimeter on the ACV range (must be on AC volts). The Video (2) board (page 142) has transformer secondaries on CB351, CB352 & Video (3) C351 (also on page 142).

Changing subjects, I assume the amplifier is still disassembled?
I had a think about this apparent loss of voltage to Q1096. In addition to R1009 (which frustratingly is OK & you confirmed it is definitely physically connected through to C1105), there's a couple of other possibilities:
1. R1280 - a 5K6 resistor on Q1096's collector, could you check if it's open circuit? If it's open, Q1096 won't get power & it also won't turn on Q1009.
2. Q1009 - check for the base - emitter diode (& might as well check for base - collector diode). If B-E open same story Q1096 won't get enough power.

Hopefully, Murphy will be kind... I'm off to bed.

[NHcoder]

I had the same problem with my Yamaha receiver.

Unfortunately my patience was less and I decided to break the problem into two logical parts.

(Attachment Link)

Obviously there is more to this story than just being clumsy. I just gave up on Yamaha amplifiers and switched to a different brand.

I will use the Amplifer and Power section somewhere else.

Haha ok.  I don't buy them often enough to have any brand loyalty, but I have had this one for ten years or more and it's worked great the entire time until very recently.  One could say I got my money's worth out of it, but I don't think that's a reason to at least not *try* to revive it.  I have a soundbar w/ satellites & a sub filling in now and while it works, it's just "ok" and no match for a real receiver.

I'm back from the bushwalk so I'm a bit stuffed but I did think of a few things.

Great, hope you had a good time.

Firstly, in relation to turning on the power transformer:

I then tried activating the transistor Q2004 with 3.3v through a 1n4001 on the PRY pin.  This pin says it takes 2.7, which itself comes from a diode at the 3.3v output of an IC on another board.  I figure the 1n4001 would be ok as it has a drop of about 1.1v, but maybe 2.2 isn't quite high enough.  I was hoping to hear an audible click from the relay that Q2004 is controlling, which activates the main power transformer, but got nothing.

You could put 3.3V or more (see below) on the Power (1) PRY pin (W2505) as long as the cable is disconnected from the connector (so it can't back feed to whatever is driving it). Q2004 base current is limited by R2050 (page 139) i.e. it's not critical although it's best not to go lower than the original 2.7V or higher than say 5.5V. The Power (1) board has got mains potential so please be careful.

This is exactly what I have been trying to do and it just doesn't seem to come on.  I don't hear a click and am not getting AC voltage on the transformer outputs as measured @ the 6 pin connector as described on page 138 for the MAIN(3) board. 

To check that RL201 is operating, I can think of a couple of options:

1. You could measure Q2004's collector voltage which would also indirectly measure the relay coil voltage.
With PRY 0V, the Q2004 will be off & its collector voltage should be 5.5V (pulled up by the relay coil) & since both RL201's coil connections are at 5.5V, it will be de-energised (no current flowing through the coil). With PRY 3.3V or higher, it's collector voltage should drop to 0V i.e. one connection of RL201's coil will be at 5.5V and the other connection (Q2004's collector) will be 0V i.e. in theory, the relay will be energised.

2. Find a reasonably safe low voltage secondary and check it with a multimeter on the ACV range (must be on AC volts). The Video (2) board (page 142) has transformer secondaries on CB351, CB352 & Video (3) C351 (also on page 142).

The power board is one I have left mounted in the receiver this entire time but with it mounted I don't have access to the area where the relay or transistor are located.  I am pulling that board out now and intend to just test the relay functionality with my meter and bench supply to make sure it's working.  If it is, I'll investigate the board itself more closely, check the transistor, etc.

Changing subjects, I assume the amplifier is still disassembled?

Indeed.  The top four boards in the stack are entirely removed and sitting in a box.  All that remains in the chassis right now are the transformer, the two POWER boards, and  MAIN 1, 2, and 3.  The front panel control/digital boards are still there but entirely disconnected except for a small board I haven't identified that is hard wired into the transformer output.  Pretty much everything that can be disconnected is disconnected, including all the connections to the digital boards you mentioned above.  Those boards are safely in a pile in a box across the room

I had a think about this apparent loss of voltage to Q1096. In addition to R1009 (which frustratingly is OK & you confirmed it is definitely physically connected through to C1105), there's a couple of other possibilities:
1. R1280 - a 5K6 resistor on Q1096's collector, could you check if it's open circuit? If it's open, Q1096 won't get power & it also won't turn on Q1009.
2. Q1009 - check for the base - emitter diode (& might as well check for base - collector diode). If B-E open same story Q1096 won't get enough power.

I'll check all of this as soon as I do my quick tests on the relay and Q2004, good ideas.

Hopefully, Murphy will be kind... I'm off to bed.

All evidence to the contrary so far   Thanks for the reply, maybe you'll awake with a eureka! moment.

[NHcoder]

Ok, the relay is working fine and clicks loudly when I apply 5.5v from my bench supply.  I had my meter's continuity check connected to the other side to make sure there was actually an electrical connection being made as well.

Q2004's diode checks were also fine, 0.7v on BC and BE.  The emitter-collector path also has a 0.7v drop which worried me for a minute, but the schematic shows it's probably fine.  It looks like a path from Q2004's emitter to the ground plane, up through D2015 and then Q2003's BC junction and back into the relay coil and thus the collector via the R2538 shunt resistor.

The protection fuse to the transformer is ok.

I had a think about this apparent loss of voltage to Q1096. In addition to R1009 (which frustratingly is OK & you confirmed it is definitely physically connected through to C1105), there's a couple of other possibilities:
1. R1280 - a 5K6 resistor on Q1096's collector, could you check if it's open circuit? If it's open, Q1096 won't get power & it also won't turn on Q1009.

"Unforutnately" tested fine, 5.54k.

2. Q1009 - check for the base - emitter diode (& might as well check for base - collector diode). If B-E open same story Q1096 won't get enough power.

Base-emitter was ok, 0.6v, but base-collector is strange, showing about 3v.  After checking this I realized that plugging the receiver back in and attempting to get the relay to switch on, I had recharged those massive caps, and they are probably interfering with this reading.  The test points for discharging them are on the bottom of the PCB, let me know if I need to pull it back out to discharge the caps and check this again.

I'm going to take another break from this now and spend time thinking about something else.  After that I'll be back to trying to convince the relay to turn on via the W2001 connector to check the transformer output unless you or someone else has a better idea.  I suspect the transformer is fine, but testing in stages from the source like this without all the other boards connected and in the way seems like it would be beneficial.