Yamaha RX-V390 - Circuit Protection

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I think you may have to force your amp to stay on in as safe a manner as possible to get more sensible readings.

If you can disconnect the bridge +/- outputs, 2 resistors in series with the DC rails would make it unlikely you would blow anything up. It would be even better if you could put them in the +-47v lines after the reservoir caps c140 c141 but I expect you would have to cut tracks to do that. Try two 100R but if they smoke you'll need higher wattage ones or lower values but then there's more chance of blow ups. Even 22R would offer a decent amount of protection.

You could force the mains relay RY102 permanently on by shorting Q122 collector/emitter (or C145).

I think CB101 unplugged is OK.
 
Last edited:
Hey guys,

Thanks for you replies!

I have remeasured from Q110/Q112/Q109/Q111 back and have found the voltages on the attached diagram. To me it seems that there is the expected voltage drops across each of the transistors so they're not SC, however you guys would know better! Note this is still with the bridge connected, I just wanted to clarify exactly what you mean first before disconnecting it!

Should I desolder the bridge and then place 2 100R in series between the -47v and 47v connectors, effectively joining the + and - rails? Then, short out the relay to stay open? What would this achieve? Wouldn't there be no power to the circuit then or only what is going through D108? Or have I misunderstood?

Cheers,
Felix
 

Attachments

  • amp2.png
    amp2.png
    190.6 KB · Views: 95
You've misunderstood, I meant break the connection at +47v and insert one resistor across that break so that all power to the amp has to go through that resistor, ditto with the -ve side, purely to limit maximum current availability. It would be best done after the reservoir caps if possible, if not then at the bridge. Not essential if you're careful, but something that can save a lot of heartache.

Your latest voltage measurements seem to show that the output stages are OK as the bias isn't far off what you may expect, about 2v across all 4 b/e junctions. Q105 doesn't look like it's conducting due to insufficient b/e voltage. I think the 37v at the output is due to something being off at the input end, so a lot more reading round there could help. Specifically, all the Q101 Q103 voltages. I presume there's about 15v at the left end of R109, that's common to both input stages and could shift the output voltage if it's way out.
 
Last edited:
Hey guys,

That makes so much more sense sbrads, I don't know why I thought that ha.

Anyway, I've gone and measured the input stages to see what you guys think. If nothing comes of this then i'll look at chucking in those resistors and permanently powering it to get some better measurements. I've been trying to get the best readings possible, however strangely after turning it on and off enough times it slowly drops from 31.2v down to around 28 then magically comes back up again....strange?

Once it starts dropping I've been measuring across points as well which tells me if they are the same, so by doing this I think they should be more accurate
:)

I tested Q105 and Q106 and they aren't shorted.

My conclusions from this (every component is reflected on the R channel also):
1. The voltages on bases of Q101/Q103 mean they aren't turning on. This then means that there is no voltage drop across R107 which means Q105 won't turn on either.
2. I have 0v drop between the TP101 rail and the base of the Q101, which initially made me think that C103 was S/C, but it isn't. (or is this to be expected?)

Don't know if there is anything in this either but I noticed if I turn on (and it shuts off) 3 times in a row quickly then RY101 clicks on then turns off with everything else.

Thanks!
Felix
 

Attachments

  • amp3.png
    amp3.png
    109.9 KB · Views: 87
Hi,
This is the way I see it. Transistors Q105 and Q107 need to conduct to balance the bases of transistors Q111 and Q109. This will bias the bases to -1.1 as shown in the schematic. If not then it can go + 37 or -37 depending of the problem. Q105 need to be conducting to the amplifier to work. Check the collector voltage of Q105 and it should be negative. I know you check it for short but it can be open too collector to emitter.
 
2 amps doing the same thing though, it's weird, so back to basics. All these type of amps have overall A.C. gain determined by 2 resistors ratio (R117, R115) and their DC isolation is provided by one capacitor (C109), so that the DC gain can be unity, set by 100% DC feedback through one resistor (R117). There is another low value resistor R153 (just off your diagram) that connects the output to R117 that I'm ignoring.

This DC feedback means that the output will be at 0v (desired!) if the input (base of Q101) is at 0v, more or less. Most circuits will have this base pulled down to 0v via a medium value (10k or so) resistor to make sure it's near 0v. Now, during your fault finding you have CB101 unplugged, so an external ground on the E pin won't be there. I'm therefore intrigued how you have been measuring 0v at the bottom of R101 (THE important place for there to be 0v) as I can't see where the 0v gets there from looking at the circuit. If that can float up, the output will go up with it.

What with the schematic errors, I could easily be way off the mark here LOL.
 
Hey guys,

Tauro - Q105 Collector has 38.2V on it, meaning it's not conducting!

I took your advice sbrads and placed two 220R (it was all I could find in my parts box!) resistors in series with the +-47 rails before the power caps as I couldn't find an easy place to put them. This has causes me to have +-42V straight of the bridge as opposed to +-47V, is that to be expected?

I have then shorted Q122 so the amp stays powered on and have found something interesting I think...

As in you noted in your previous post sbrads, where is the 0v connection on the inputs? When I first starting measuring I had 32.2V on TP101 (and all connections that are in parallel with it), however this steadily started dropping with my multimeter connected. It got all the way to 19V when I powered the amp off however I am sure it kept going.

My take on this: C109/C110 are powering this up until those caps become charged. I measured one the + side of these caps, the top side of C105, C/E of Q101 and I have a constant and stable voltage there.

So I think CB101 must have a ground connection somewhere on the other board that this needs.


Thoughts?
 
42v is fine for safer fault finding. It also means your amp isn't drawing loads of current which is also good.

It's looking more and more like you have a 0v connection problem. If you have another 220R resistor (again, for safety rather than a dead short!), connect it between the 0v right at the reservoir caps and the bottom of R101. See if that amp's output voltage drops to near 0v. If that works, try the same on the other amp. R101 and R102 are a bit high really, so the output may still have a small offset voltage.
 
Last edited:
Hey guys,

Finally got around to getting some time to do some more testing, I grounded a 220R on the bottom of R101 and this is what I have found (attached).

My observations:
1. I think Q101 is now conducting as I now have a voltage drop across R107 and I have 0.76V on both sides of R109, (it was 14.4V on the left and 17.2 on the right when I didn't ground out R101)
2. I have a stable 26.5V at the output of the left channel and 26.45V at the output of the right channel. Measuring across the outputs from L to R I have 50mV...
3. I still have 0.7V at R165 meaning protection still kicks in.
4. It's still mirrored (my measured voltages from the diagram) onto the R channel as well however measuring from the RHS of R119 to R120 I have 58mV. This could just be because of component tolerances I guess...
5. I have the expected voltage drop still across the B/E of all the transistors, so I'm thinking it's not one of those?

What do you guys think? Should this be grounded through CB101 normally or?
I'm really not coming up with much else... :(

Thanks again!

Felix
 

Attachments

  • Amp4.png
    Amp4.png
    117.5 KB · Views: 72
Hey Guys,

Sbrads I think you are onto it, looking around D109 I had 0.75V when I had R101 grounded, however 14.5V when it wasn't. I then noticed the top of the Q133 FET has a 25v supply, coming from the rear power amp. I plugged that board back in and it gave me back my 14.5V and a lot more sensible voltages on all of the transistors (-1.1 on Q105 for example). So, I think it turns out this does have to be connected however I now have 150mv on both channels at the output and -0.7V at R165 meaning circuit protection still kicks in. Now I am getting proper readings though I will be able to do some better diagnostics!

Regarding CB101, I'm pretty sure the +-15v there are supplying the other board, however the earth connectors I think are actually needed. It would be a little tricky to plug this back in and still easily test, so would it be safe to scrap the 220R (going to ground from R101) and directly wire the earth pins of CB101 to 0V? This way I wouldn't have the 500mV around there and I can get precise readings... I still have the 220R in with the bridge so I shouldn't be able to draw too much current... thoughts?

I should hopefully get some more time tomorrow to get some better readings!

Thanks again!
Felix
 
Hey Guys,

Sbrads I think you are onto it, looking around D109 I had 0.75V when I had R101 grounded, however 14.5V when it wasn't. I then noticed the top of the Q133 FET has a 25v supply, coming from the rear power amp. I plugged that board back in and it gave me back my 14.5V and a lot more sensible voltages on all of the transistors (-1.1 on Q105 for example). So, I think it turns out this does have to be connected however I now have 150mv on both channels at the output and -0.7V at R165 meaning circuit protection still kicks in. Now I am getting proper readings though I will be able to do some better diagnostics!

Regarding CB101, I'm pretty sure the +-15v there are supplying the other board, however the earth connectors I think are actually needed. It would be a little tricky to plug this back in and still easily test, so would it be safe to scrap the 220R (going to ground from R101) and directly wire the earth pins of CB101 to 0V? This way I wouldn't have the 500mV around there and I can get precise readings... I still have the 220R in with the bridge so I shouldn't be able to draw too much current... thoughts?

I should hopefully get some more time tomorrow to get some better readings!

Thanks again!
Felix

It should be safe to link 0v to CB101 E pin, but I always believe in safety first so I would be tempted to put 10R in series, measure the voltage across it and if the current (ohm's law and all that) is sensible then you can short it out. Sensible current? Experience is everything there, but <100mA should be OK and I would expect it to be much less than that.
 
Hey Guys,

Did as you said sbrads and only had around 1mA, so straight grounded it and everything is fine. Getting MUCH more sensible voltage readings now. (Finally, it feels like we're getting somewhere!).

So, looking at my new measurements I have around -29mV on the output of the left channel and -27.2mV on the output of the right. Is this too much negative bias? Measuring all the junctions of the transistors I have the correct voltages as I should (I think) except for this smallish negative bias.

Looking at R165 I still have -0.74V, so protection still kicks in. How can this be? Is this because of the 1.07V I have on the bottom of D103? (the muting circuit being fed from the IC?).

Let me know what you guys think, nothing obvious is hitting me...

Cheers!
Felix
 

Attachments

  • Amp.png
    Amp.png
    126.4 KB · Views: 60
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.