AVS Forum | Home Theater Discussions And Reviews (https://www.avsforum.com/forum/)
-   Ultra Hi-End HT Gear ($20,000+) (https://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/)
-   -   Lexicon MC-12 repair (PS and left side surround) (https://www.avsforum.com/forum/86-ultra-hi-end-ht-gear-20-000/3080738-lexicon-mc-12-repair-ps-left-side-surround.html)

sfogg 07-25-2019 09:29 PM

Lexicon MC-12 repair (PS and left side surround)
 
WARNING: HERE THERE BE DRAGONS! Proceed at your own risk.

Just posting this as info if anyone else needs it. Do not try this at home. Proceed at your own risk. You break it you bought it...etc...etc...etc.

Background: My MC-12HDB developed a problem with the left side surround not working years ago. Pyramid Audio repaired it but later on the same channel developed some low level noises and such.

This past weekend the power supply in the Lexicon died. Mine had the newer EOS LFVLT110-4302 in it and I had purchased a spare years ago. The EOS supply will also work in MC-8s and MC-12s that have the older power supply but you have to repin the connector inside the Lexicon. Since I was replacing like for like I didn't have to do that.

The EOS supply is actually kind of a pain to get out as access to two of the screws isn't there. It looks like it was bolted to the interior PS panel before the whole chassis was put together. You can remove the interior PS panel to get it out but it involves a fair number of screws and if you have a balanced Lexicon you have to remove the balanced chassis section to get at some of them.

I went a different route and added 4 access holes to to the interior panel to be able to get at the screws easier. Should have drilled the holes a step larger so that the screw heads would fit through the holes too.

https://live.staticflickr.com/65535/...353589bc_c.jpg

If you do this drill very slowly (to contain any metal shavings) and be sure to completely clean up the power supply section to remove any possible metal bits. Deburr the inside too.

The older style power supply was held on by two screws accessible from the interior of the unit so they are easier to remove.

I was expecting to see some dead caps on the dead supply but it really burned itself up.

https://live.staticflickr.com/65535/...1d98ac6b_c.jpg

After getting the new supply installed the Lexicon was back from the dead.

At this point I decided to tackle the left side surround problem.

The Problem:

First, I had to determine where the noise was coming from. The left side surround circuit has the digital to analog section and then it feeds into the volume control and output section. I could see Pyramids repair was in the DAC section but I wasn't sure if that was where my noise was getting in too.

Easy way to pin point this is to get the system in its regular processing mode and listen for the noise. I verified that I could her the noise in this state. Then I put the system in true 6channel analog bypass (Input Advanced-Analog Bypass ON) with the rear panel set for the 6 channel input. This takes the analog input and feeds it directly to the volume controls for the outputs. Once I put it in this mode the noise was gone and the channel was outputting audio fine. This told me the problem with in the DAC section of the left side surround and that the switching and volume control section of that output was fine.

If you have the same problem, left side surround is noisy or dead during processing but is OK for true 6 channel analog bypass, than this should fix your problem. If your problem is different, this will not fix it.



Design Background:


The MC-12 has one 2 channel DAC (AD1853) for each of its main channel outputs. Each dac is run in dual differential mode and all the DACs are fed from the FPGA on the analog board. The FPGA sort of unpacks the SHARCs data and creates the inverse signal for the the dual differential mode of the serial data to the DACs as well the master, bit and L/R clock to the DACs.

The MC-12 has 12 outputs in the main zone. 2 of the channels were never utilized, those are the L/R AUX channels. The circuitry is *almost* identical for all the 12 channels, except for the AUX outputs. The AUX outputs don't have the switching ahead of the volume controls to allow for the true analog bypass on those channels.

Instead of just trying to repair the digital section of the Left Side surround I decided to go a different way of fixing this. I would use the DAC section of the Left AUX output to convert the left side surround channel back to analog and then to inject that analog signal into the functioning Left Side surround volume control and output. I wanted to get back into the actual Left Side surround Volume section to keep things easier with regards to all that happens internally for channel balancing and volume control. If I used the aux output itself I'd have to trick the Lex. into treating that volume control as if it were the left side volume control. I would also loose the true analog bypass ability for that channel as the aux outputs do not have that switcihing between the DAC and volume controls.

After many hours looking over the circuit board, reading datasheets and using my multimeter to trace out a bunch of things I found a number of important details.

First, all the main channel DACs are wired together with regards to the setup and of the DACs. (Pins 3,4,5) That meant that I didn't have to do anything special to configure the Left Aux dac. The Lex would do it for me.

Second, the master, bit and L/R clocks are the same for all the main channel DACs. This was trickier as they are not all actually in continuity with each other and some are labeled as different clocks on the board itself. But that is because they were buffered independently for each dac. What feeds the buffers are the same clock signal from the FPGA. This too was good news as it meant I didn’t have to touch the clock signals on the Left Aux dac as they were already identical to the left side surround DAC.

Third, Lexicon designed their boards well. Easy to follow, consistent between channels and they put in loads of test points, which became very helpful.



Hypothesis:

After looking into all of this I decided I would try routing the left side surround serial data (the actual audio data) to the left aux dac and then taking its analog output (after filtering/buffering) and injecting that back into the switching and volume control section of the actual left side surround output. Since my noise was coming from the digital section this should get rid of it.

The tear down:

Turns out that after I planned this all out it is fairly straight forward to do. Not going to give step by steps on the chassis teardown but basically you need to remove the analog board from the Lexicon to be able to work on it. Remove all the expansion cards and then the analog board will come out pretty easily. If you have a balanced output version remove the video board as you are going to need to be able to tilt the audio board up to remove the ribbon cable (on the underside of the board) feeding the balance output section. This will also make reconnecting that ribbon cable much easier. If you don’t have balanced outputs you don’t have to remove the video board.

After you get the analog board out you will need to remove the metal shield from the bottom. Remove the sticky and likely nasty foam spacer that is under there. You will likely find it is under the left side surround DAC section. I actually think that tape is what is causing the problems on the left side surround. It seems to degrade the contacts on the vias in that area. I put a couple of hard rubbers bumpers on the shield itself to act as a spacer to prevent the board from touching the spacer in the middle.


There are two parts to this modification. The digital section and the analog section.

Digital section Modification:


For the digital section we need to reroute the left side surround serial data to the Left AUX dac.

First: We need to cut the trace from the FPGA that is feeding test point E47. This disconnects the Left Aux output serial data from the left aux DAC.

https://live.staticflickr.com/65535/...837abe48_c.jpg

I used an exacto knife to cut the trace and verified that there was no longer continuity with the FPGA. Don’t cut deep, just enough to remove this trace.

Second: We jumper from test point E52 to E47. This connects the left side surround serial data to the left aux DAC.

https://live.staticflickr.com/65535/...2ef279f0_c.jpg

Analog Section Modification:

First: we need to disconnect the left side surround DAC from the left side surround analog bypass switching and volume control. To do this we remove two components from the board. Remove R354 and C252. This is a surface mount resistor and a through hole capacitor. These are part of the filtering that occurs right before the analog test point and before the signal is put into the analog bypass switching ahead of the volume control. Resistor and capacitor are removed in this picture.

https://live.staticflickr.com/65535/...ec71310d_c.jpg


Second: We need to take the analog output from the Left Aux DAC and inject it into the left side surround analog circuitry. Do this by jumping from E21 to E32.


https://live.staticflickr.com/65535/...5dafacb8_c.jpg

That is it. Put it all back together. Make sure you reconnect all ribbon cables (even the hidden one for the balanced outputs if you have them), install the expansion cards in the right slots and on the correct sides of the mounting bars.

Done this way my left side surround output is now working properly and the system has all the capabilities it always had on that channel, including analog bypass.

Shawn

LydMekk 07-27-2019 11:47 PM

I had the same problem around 10 years ago. Kind of strange, suggest a common design flaw...mine was original an 3.00 version from 2002-2003 ish. Later upgraded to 5.00 with RC/mic board/kit.

Sent the boards in to Pyramid audio and had them fix it, exact same left side surround, level problems and noise.
Back then I managed to buy a new PS, newer version, which I installed myself together with the boards etc.
Went swimmingly, no problems. Have been running good now for 10 years since the install.

So, thumbs up for your unit also ^^.

sfogg 07-28-2019 07:37 AM

I have seen a few others that had a problem with the left side surround too. I don't think it is a circuit design flaw as the left side channel is identical to all the others and laid out the same way on the board too. I think the problem is with the emi shield and the piece of rubber/foam used as an insulator between the circuit board and the shield. Over time the foam degrades and I think it chemically reacts with the traces or vias (or maybe just the insulation) on the circuit board. On the few analog boards I have seen the foam was attached under the left side surround DAC section and after removing the foam the board looks a little altered in that area. My guess is it degraded some traces/vias enough that the clocks or serial data to the left side DAC were getting corrupted enough to cause noise issues in the DAC.

Shawn


All times are GMT -7. The time now is 05:30 AM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
vBulletin Security provided by vBSecurity (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.