Denon AVR-1912 E2/EA

[Michael Studio1]

I have adjusted the Idling Current (page 56 in Denon Service Manual) And all 7 channels adjusted to the required 3 mV. This means the new Darlington pair, its new controlling Transistor and associated 47Ω power resistors on Channel SBL are functioning OK. (phew!) So I'm back on taking Voltage measurements. First, the power supply to IC801. Pin 30 (AVCC) reads +7.416V. and Pin 52 (AVEE) reads -7.050V. I have injected two Signal Tones at VAUXL - 500cps and VAUXR - 700cps. These are both getting through IC801 at Pin 21(FLOUT) Pin23(FROUT) So I was certainly right about changing Cap C799 on account of its high ESR. Next step to find both Signal Tones at the ADConverter IC30 on HDMI B'D (Page 106/B5) YES! both tones are there. So we're in with a chance! Since I have no test instrument to check a Digital presence of these tones I must check at the DAConverter . . . which is?What? IC? I am finding this very confusing and need help. True I could test all possible pins on IC801 - but is IC801 a DAConverter? Michael Studio1 UK 17:39BST

 

[Michael Studio1]

I have erred! The four Power resistors in each of the 7Channels are 2W. 0.47Ω - NOT 47Ω! I did get the correct ones from China though - 2W. 0.47Ω - and, as I said, they're in, adjusted for Idling Current to 3mV. and working.
Now I have to find my treatise wherein I searched and found the route from DAConverter to SPKR and here it is: The DAConverter is IC29 (page126/G5) The Digital signal comes in at pin 14 (SDT11) and goes out as Analog at Pin 1 (LOUT1-) Pin 2 (LOUT1+) Pin 47 (ROUT1-) Pin 48 (ROUT1+) These four conductors (2 Left and 2 Right) go to IC28A (FL)page 126/K4 and IC28B (FR) page 126/K5 both going to DIGITAL CNT as [DAFL] and [DAFR] respectively on page 129/E3-5 on 33-pin Connector Block CN2 at pin 28(DAFL) and pin 26(DAFR) which mates with Connector Block CP11 page 112/J-K-L3 where they connect to the 17 pin CP115 at pin13 (DA_FL) and pin12 (DA_FR) Thence to CN115 page 114/F-G1 and on to IC801 pin93 (D/A_FL) and pin94 (D/A_FR). Then through IC801, emerging at pin21 (FLOUT) and pin23 (FROUT) as [AMP_FL]> and [AMP_FR]> where both go to CN401 page 114/O3-4 a 13pin Connector to the 7CH_AMP at CP401 page106/A4-5 pin1 (FL) and pin9 (FR) to feed those two Channels. I guess that following the route of these two signals to their ultimate Speaker Terminals is now straight-forward. But . . . . am I correct in my routeing? Michael Studio1 UK 13:08BST

 

[Tha fios agaibh]

Hi Michael,
I don't have time at the moment to follow along, but its just a matter of following the yellow brick road. A bit much to decipher all the details in your text and go back and forth to the schematic. I'm sure what you've found is likely correct, and makes perfect sense, at least to the person looking at the circuit.

Sounds like your making great progress. Continue following the analog path and if you have a question, posting a screen shot picture (prt screen) of the area would honestly be more helpful than a detailed text.
I remember talking about the location of the ad and da chips a while back but it sounds like your back on track.
Regards, John

 

[Michael Studio1]

Hi Michael,
I don't have time at the moment to follow along, but its just a matter of following the yellow brick road. A bit much to decipher all the details in your text and go back and forth to the schematic. I'm sure what you've found is likely correct, and makes perfect sense, at least to the person looking at the circuit.

Sounds like your making great progress. Continue following the analog path and if you have a question, posting a screen shot picture (prt screen) of the area would honestly be more helpful than a detailed text.
I remember talking about the location of the ad and da chips a while back but it sounds like your back on track.
Regards, John

Thanks, John. Sorry my text is a bit too much to decipher - really it's just an 'aide-memoire' for me - just thought I'd post it as it shows my meaningful questions. There is one thing I'm having location difficulty with - and maybe there's a short-cut way of doing this - where actually IS IC29? I've been scanning the HDMI B'D on page104 going up, going down, going sideways with 400% enlargement . . . and I just cannot find it. Did you say type in the component and hit 'go' - and it'll be there? I can't find where to type in my request for starters! Michael Studio1 UK 18:17BST

 

[Michael Studio1]

Wonderful, John. Using your method of the 'magnifying glass' in Adobe I have found IC29 in Cct.Diag. P126/G5. Now all I need to do is find it physically on the PCB. It says it's on 'MAIN DAC' Cct.Diags (HDMI UNIT 21/26) I'd like to find it on page 104, though. So I'll try this locator with page 104 selected - and see what happens! Hoo-Rah! It found it on page 105/2F. Now that IS something! All of this means I shall be able to continue probing components for the Signal Tones. Michael Studio1 UK 18:59BST (British Shivering Time - it's +4 degrees here.)

 

[Michael Studio1]

It's no wonder that I couldn't find IC29 on the HDMI_B'D - because it's NOT on the 'Component' side but on the 'Foil' side. I found it using the Endoscope. Michael Studio1 UK 19:26BST

 

[Michael Studio1]

Glad you liked this, Dries. Now I have gone further and powered up the AVR, applied Signal Tone to VAUXL & VAUXR. Caramba! The tone appears at SPKR terminal FR (but not at FL). More probing and testing to do then. I have managed to download and mess around (size-wise) with the Denon Service Manual pages on the Studio DELL'Chrome' which has enabled Screen Shots and Printing. This, in turn, means I have larger Cct.Diags. and pcb printouts pegged up on a couple of Music stands in the Lab allowing me to trace the Signal for VAUXL and see where it disappears. I have enlarged areas of the HDMI B'D shewing IC29 page 126/G5 and IC28 p.126/J4&5. With the help of the Foil prints page 105 and the Endoscope on the Lab PC I shall be able to place the probes on the Foil side of the HDMI B'D to access the contacts of IC29 p.105/G2 and IC27 page 105/H1. These are difficult to see on the Foil side without enlargement - so that's what I have done. Michael Studio1 UK 15:37BST on 1st May (bring out your May Poles and start dancing!)

 

[Michael Studio1]

A little later. I have probed and found both Signal Tones on the CX4 Ribbon Cable - which connects the VAUXL&R Jacks in the Front Panel to the HDMI_B'D. This is all on Page 129/N6-9. Tracing these from CX4 pin6 (VAUXR) and pin8 (VAUXL) to Connector Block CN2 p.129/H2-5, I find the Signal at pin4 (VAUXR) but NOT at pin6 (VAUXL). So somewhere between these two points VAUXL Signal Tone goes missing. I will re-check this bold statement . . . . Yes. No doubt about it. And there's not much by way of components between these two points - just two pairs of coils (one at each end of each of the two conductors) labelled: 'FB89' and 'FB51' for VAUXR and: 'FB91' and 'FB54' for VAUXL. I think it is FB and not F8? In all cases and at both ends of this connection the coils have the same designation: '0(1608). So it's all happening on Page 129. Interesting. Michael Studio1 UK 16:45BST

 

[bushtech]

FB =Ferrite bead? An RF suppressor

 

[Tha fios agaibh]

So the path is from CX4 pin 8, Fb91, to, Fb54, Cn2 pin 6. Correct?
I would power down and check for continuity between these points. The inductors seldom go open circuit. More likely it is a cold solder joint, or the Cx or Cn connections are bad.
Try pushing down on the connections to see if you can pinpoint a bad connection.
Alternatively, if your very careful (and not had too much coffee from your fancy machine) you could try jumpering out (bypassing) these connections to see if it comes back to life.

Excellent progress Michael.

 

[Michael Studio1]

FB =Ferrite bead? An RF suppressor

So the path is from CX4 pin 8, Fb91, to, Fb54, Cn2 pin 6. Correct?
I would power down and check for continuity between these points. The inductors seldom go open circuit. More likely it is a cold solder joint, or the Cx or Cn connections are bad.
Try pushing down on the connections to see if you can pinpoint a bad connection.
Alternatively, if your very careful (and not had too much coffee from your fancy machine) you could try jumpering out (bypassing) these connections to see if it comes back to life.

Excellent progress Michael.

I am sure you are right, John. It seems every time I turn it off, there's a new problem so soon as I power it up again. Which is why I had another 'go' at checking that fault. Dries suggests a 'ferrite bead' I can't say I know about these - or whether they are likely to exhibit an 'open Cct. I'll go for the possibility of a cold solder joint - knowing the history of this AVR. Originally when powered up the Left Chennel wouldn't work until the PCB stack was physically rotated. That truly smacks of a dry solder joint. Maybe I should apply the hot air soldering gun to it using liquid flux to restore the solder connections. I used to do this with MoBo's and had a certain amount of success - but then I was using a hot air paint stripper! Now things are a bit better controlled with the 858D soldering station. Coffee. Now that's a good notion - but it's too late in the day to fire up my fancy Italian chromed La Pavoni Europiccola with its hot steam nozzle for frothing the milk. No sugar - it's bad for you. Michael Studio1 UK 20:08BST.

 

[Michael Studio1]

FB =Ferrite bead? An RF suppressor

That seems very likely, Dries. Since it has been suggested there could be a dry-solder problem I am going to apply my 858D Controllable Hot Air Soldering tool to the CN2 and CX4 connections - see if that solidifies things. But B4 I do that I will, since I now have the HDMI_B'D out of the Stack, do some continuity tests. Michael Studio1 UK 10:48BST

 

[Michael Studio1]

Using the Service Manual I have been identifying the paths of the VAUXL and VAUXR through the AVR. This is to give me a 'tick-list' and keep my brain on track. But, with that wretched PCB stack it's very difficult to access all the test points. The two Signal Tones injected at AUXL and AUXR can only be traced 'so far' and, unlike the other day when the Right Signal Tone emerged at SPKR FR - now it isn't emerging anywhere. My latest test showed that the two signal paths are now OK where yesterday there had been an 'open' in the Left path. Just to make sure of that I have taken a mini soldering iron and have re-flowed just the few points where they had been questionable. Now the signal tones get through to CN2 - but I cannot trace them further. For this I shall have to devise and instal more 'Test Tails'. I must 'sleep on it' as they say. Michael Studio1 UK 19:19BST

 

[Tha fios agaibh]

It may be that the problem lies with the ribbon cabling or plug and socket connections. You have already found one bad cable, so chances are good there may be another. Your symptoms (intermittent problems) are also indicative of a loose connection. When you disturb the cables or plugs that interconnect the boards you either solve or create a problem.

You could try unplugging each cable and look very closely for any pins that may be pushed back slightly or have corrosion that lends to a bad connection. Also do an end to end continuity test while flexing the ends to see if they pass inspection.
Remember that you may have to test at nearby components such as the ferrite beads when necessary.

Edit: Good chance that the ustream cn/cp 103 is dropping the signal.

 

[Michael Studio1]

It may be that the problem lies with the ribbon cabling or plug and socket connections. You have already found one bad cable, so chances are good there may be another. Your symptoms (intermittent problems) are also indicative of a loose connection. When you disturb the cables or plugs that interconnect the boards you either solve or create a problem.

You could try unplugging each cable and look very closely for any pins that may be pushed back slightly or have corrosion that lends to a bad connection. Also do an end to end continuity test while flexing the ends to see if they pass inspection.
Remember that you may have to test at nearby components such as the ferrite beads when necessary.

Edit: Good chance that the ustream cn/cp 103 is dropping the signal.

Thanks, John. I haven't had the Front PCB apart so haven't disturbed CN/CP103. The signal is at all times now, getting through to the HDMI_B'D via CX4 - it's just not arriving on either L or R at the two caps. I replaced on the HDMI_B'D on page 126/K2 & K3, C799 (Right) and C831 (Left) on their way to IC30 (B5-6) to be digitised. I think there are gremlins at work. Or Trolls - whatever. What I need is a Green Man to chase them away!! Michael Studio1 UK 21:59BST

 

[Tha fios agaibh]

I'm not sure I follow. I thought we were looking at the path of VAUX. The caps C799, C831 are in the ADIN path. The source of "A/D IN" originates at IC801 pin 54 then goes through CP/CN115, CP11 and then goes through your replaced caps.
I'm sure pinpointing the signal is difficult to say the least, but lets make sure we are on the right path. Literally!

Lol, John aka, Green Guy.

 

[Michael Studio1]

I'm not sure I follow. I thought we were looking at the path of VAUX. The caps C799, C831 are in the ADIN path. The source of "A/D IN" originates at IC801 pin 54 then goes through CP/CN115, CP11 and then goes through your replaced caps.
I'm sure pinpointing the signal is difficult to say the least, but lets make sure we are on the right path. Literally!

Lol, John aka, Green Guy.

 

[Michael Studio1]

I'm not sure I follow. I thought we were looking at the path of VAUX. The caps C799, C831 are in the ADIN path. The source of "A/D IN" originates at IC801 pin 54 then goes through CP/CN115, CP11 and then goes through your replaced caps.
I'm sure pinpointing the signal is difficult to say the least, but lets make sure we are on the right path. Literally!

Lol, John aka, Green Guy.

Yes, John, that is so - but it is the very last point on both VAUXL and VAUXR that the analog signal can be heard. After this both go into IC30 where they are digitised. Previously I have heard the Signal Tones injected in VAUXL and VAUXR at both C799 and C831 so, to me, the non-appearance at these ELECT. Caps is indicative of a problem in the VAUX path. ADIN I interpret as Analog/Digital IN (to IC30) i.e. the same path with a different name. What do you think to this? I haven't spent any time on the AVR today for a 'new' Electro-Voice DO54 microphone arrived for testing - an eBay purchase which failed miserably on test and has to be returned to the seller. Michael Studio1 UK 20:23BST

 

[Tha fios agaibh]

Yes, John, that is so -...
Previously I have heard the Signal Tones injected in VAUXL and VAUXR at both C799 and C831 so, to me, the non-appearance at these ELECT. Caps is indicative of a problem in the VAUX path. ADIN I interpret as Analog/Digital IN (to IC30) i.e. the same path with a different name. What do you think to this?

Yes, could be. I'd have to take a closer look, but I'd be cautious that there isn't a process or sequence (of an IC) along the way that is inhibiting the signal from passing.
When it boots up, is it in the correct mode to pass the VAUX all the way through?
Personally, I would try to stay with a generic analog audio input (such as RCA jacks on back panel) to inject your signal, and make sure its in an analog mode such as "Bypass", where you know it should pass it all the way on to the final amp without needing to lock onto a digital signal.

BTW, If you were able to hear your signal amplified through your speakers, it'd also be a great confidence builder.

 

[Michael Studio1]

Yes, could be. I'd have to take a closer look, but I'd be cautious that there isn't a process or sequence (of an IC) along the way that is inhibiting the signal from passing.
When it boots up, is it in the correct mode to pass the VAUX all the way through?
Personally, I would try to stay with a generic analog audio input (such as RCA jacks on back panel) to inject your signal, and make sure its in an analog mode such as "Bypass", where you know it should pass it all the way on to the final amp without needing to lock onto a digital signal.

BTW, If you were able to hear your signal amplified through your speakers, it'd also be a great confidence builder.

Thanks John! You give me ideas I had forgotten about - the use of other Analog Inputs at the back. I had used the VAUX because it is there 'by default' when the AVR is powered up. I thought the Signal Path was switched in the Digital area and that those C799 and C831 were specifically in the VAUX path. What had happened was, following the great 'tear-down' and Electrolytic Tests on all PCB's when put together the VAUXR signal got through to the FR again and was present at C799 and C831 - so it was somewhat depressing that everything should 'implode' again. Now, before I do anything further, I shall power up the AVR and inject those signals into the DVD RCA sockets at the back, select DVD and see what I get at both C799, C831 and FR/FL speaker terminals. Fingers X'd! Michael Studio1 UK 08:03BST