The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Synths for sale  Search Gearslutz Go Advanced
Troubleshooting an Alesis A6 Andromeda - No Boot
Old 5th February 2015
  #1
Gear Addict
 
M.Retra's Avatar
Troubleshooting an Alesis A6 Andromeda - No Boot

Hello folks! For those A6 gurus out there, I need some help. I am troubleshooting a "somewhat lifeless" A6. I say somewhat because the unit powers up but does not display a splash screen.

So far, I have obtained a lot of information regarding the operation of several IC's, ports, pins, and interfaces inside the A6. As I have experience in board-level analysis, and tools and test equipment to allow for repairs and troubleshooting.

The person that I purchased it from said that it worked well a month or two before the sale. So I am holding onto hope.

Most things seem to be in order, hardware wise. The power supply seems OK; all voltages seem good, not only at the PSU terminals, but all around the boards at various points as well. Crystal oscillators are all oscillating at their respective frequencies. The 87C52 is sending out pulse trains on the read pin and the LCD cable pins, which lends me to believe the 87C52 and the Front Panel Left PCB should be OK. But the fact that absolutely nothing is displayed on the screen is the weird part. At this time, I cannot confirm if the LCD is 100% functional as I don't have a spare.

As far as I can tell the Front Panel Left PCB EPROM chip should be programmed correctly. How can I confirm this? Well, with experience in programming 87/89C51 micros myself, it's either all or nothing. If the device is not programmed correctly internally, or did not receive the proper program from an EPROM IC, then it's not going to work, period. So if the EPROM IC is removed, and the unit started up, no pulse-trains occur on the LCD cable. However, upon reinsertion of the EPROM and subsequent power up, then the pulse trains are back. That, to me, confirms the programming of the EPROM to be OK.

Other than that, I found some easy fixes such as two, bulged and leaking capacitors on the main output. Yay for easy fixes!

Alas, I need to ask a couple of basic questions and confirm a few things:

1. The stupid question: what is the sequence of events that occurs when a normal Andy is powered up? Since I have no experience in the Andy, I don't know what to expect. So if the screen is indeed dead, then perhaps it would display something if it could. Know what I mean? I've searched for videos on YouTube, but nothing showing what I need. Is there any sort of confirmation button I need to press, like "OK to proceed," or does it simply show the splash screen, light up the LEDs, and change the display to EDIT (or something), and allow keys to be pressed and make sound?

2. Per the Service Manual, the 87C52 processor handles the LCD. So does that mean all of the graphics are handled by the 87C52 too? If so, then in theory, shouldn't the 87C52 processor be able to display something without being connected to the Main Board? This really isn't possible since the main ribbon cable provides power.

3. The observed pulse-trains on the Coldfire processor, as well as the various flash IC's, are a little "iffy." I'm trying to look for periodic pulse-trains that would confirm communication between devices. Right now, it's a toss up. I don't suspect a possible bad Coldfire "Host" CPU, but I do suspect a possible flash or programming issue. Therefore, how likely is this to occur? Have any of you with trouble Andy's had the flash IC replaced and/or reprogrammed?

4. The Service Manual states the SRAM IC's could be problematic. Those may be easy to replace as they don't have programmed data on them. But how many Andies out there had their SRAM chips replaced? If many users report that their SRAM chips have been replaced, then it would become a strong consideration to swap them out.

5. From what I can gather from the Service Manual, "ECN's" (Engineering Change Notices) have occurred over the years to update the boards, replace components, etc. Think of these like factory TSB's for your car. However, what ECN's are out there? What was involved?

As you can see, I'm getting pretty deep in repair of the fragile monster known as the Andromeda. I thank you all in advance for any information you can provide. And I believe that every thread should have a resolution and closure, so I will keep working on this thing and not give up!

If I truly exhaust all my skills and abilities, I will take it to a repair shop and give them the dossier of what I've done up to that point!
Old 5th February 2015
  #2
Lives for gear
 
CthonicEwes's Avatar
You seem very technically adept. All I can say is that if your main PCB is fried you are out of luck. Replacements are unobtanium. There is a guy in LA who works on these things. I talked to him. He said there is no way to fix a busted main PCB. I know it's not much help...good luck.

Sorry, I forgot the guys name, but if want me to do some research, let me know. Also, there's some guy in NYC who works on the Alesis A6 and he is to be avoided at all costs.

Oh, yeah, and the really stupid sugggestion: did you mess with the screen contrast?
Old 5th February 2015
  #3
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by CthonicEwes View Post
You seem very technically adept. All I can say is that if your main PCB is fried you are out of luck. Replacements are unobtanium. There is a guy in LA who works on these things. I talked to him. He said there is no way to fix a busted main PCB. I know it's not much help...good luck.

Sorry, I forgot the guys name, but if want me to do some research, let me know. Also, there's some guy in NYC who works on the Alesis A6 and he is to be avoided at all costs.

Oh, yeah, and the really stupid sugggestion: did you mess with the screen contrast?
Yes, I checked the voltage on the IDE connector that goes into the LCD module. When the pot is turned, the voltage does indeed change. I mainly looked for a range of voltage rather than a specific voltage. So at this time, it appeared OK.

As for the LA repair guy, if you recall the name, then let me know. It's worth at least a discussion.

Overall, the Andromeda is a scary synthesizer. The entire situation is scary. Alesis should have realized that early on. There is truly too much that can go wrong in the thing; it's too complex and relies on too many proprietary devices, too many communication channels, and code.

Overall, I took a chance on the purchase of the non-working synth. At this time, I'm not 100% convinced that the Main PCB is fried. I'm getting a lot of signs it may still be alive. But if it comes down to merely being a bad flash (programmed) chip, then that's the Achille's Heel, which is a really bummer way to go. At the very least, if all else fails, then it will be sold for parts.

I find it strange that the Service Manual states the following and then basically gives up:

3.201 Unit Doesn’t Boot – No Splash Screen

• If the backlight doesn’t go on, check 5V power on the top panel.
N/A as the backlight comes on.
• Check the ribbon cable going from the Front Panel Left board to the LCD.
Seems OK as pulse-trains and voltage are all present.
• If the backlight goes on, but there is no splash screen, make sure the Front Panel EPROM is properly programmed and seated in its socket (page 4 of the Front Panel Left schematic).
No way to know for sure, but it appears to be programmed and seated properly because I am seeing "action" on the 87C52 when plugged in and no action when not.
• Make sure crystal for the 8052 is oscillating, and that the proper RD, WR, etc. signals are being generated.
Same as above. I am indeed seeing oscillation on the crystal and "action" on the RD and WR pins.
• Make sure the LCD contrast is turned to a good value, and make sure the LCD is working on another unit.
I assume it is set at a good value. However, no change occurs on the LCD despite twisting the knob full CW or full CCW.

But that's where their troubleshooting ends!?? That's it? What if the LCD module tests OK?

As far as I can tell based on the block diagram in the Service Manual, the data connection to the LCD Module is a one-way street. Basically, like most LCD modules, it is expected to respond and display data without giving feedback. That may not be the case with the Alesis-spec'd module, however. I need to investigate further. If the data path is not a one-way street, then my LCD may not be talking back (hand-shaking) to the Front Panel Left PCB, then in turn, pushing that hand-shake agreement back to the host controller, i.e. the Coldfire CPU. I am really hoping this is the case. A $200 replacement of the LCD is certainly better than a giant paperweight!

Again, it is very strange that the LCD doesn't even blip. Absolutely nothing is displayed on the screen despite turning off/on several times. One would think at least something would pop up on the display!
Old 5th February 2015
  #4
Lives for gear
 

Seems like threads like these are coming up weekly now. Andromedas dying everywhere. Not booting or not booting fully. The Flash on the main board is probably dying or dead on all of these but it's not easy to replace.

My Andromeda is at my techs. I can get mine to get to the splash screen and the physical part of the synth works but it won't go over to the actual synth.

Same thing. Bought mine with some quirks and then went to turn it on one day and it wont go past the splash screen.

In the old days I'm pretty sure Alesis just pulled out the main board tossed it in the garbage and threw a new main board in. But now there aren't any more main boards. And Alesis isn't making more. So my Andromeda just sits at my techs. It doesnt seem to have user replaceable parts. I havent heard of people fixing the main boards on their own.

I hold out hope that some 3rd party will make new boards one day and all of these Andromedas will come back to life. But this seems to be a common occurence now. Had I known this before I would never have bought one. It would be nice if one of the original engineers could help the cause.

While it worked I was happy. Now I'm just pissed off. And a bit sad more than anything.

Last edited by plikestechno; 5th February 2015 at 09:07 PM..
Old 5th February 2015
  #5
Lives for gear
 
CthonicEwes's Avatar
So sorry to hear about that, plikestechno.

The only positive thing I can add is that it doesn't sound like M. Retra has the hanging flash screen indicative of a bad Main PCB board. I'll do some research on the A6 guy in LA. I talked to him about a year ago, but I don't recall his name or the name of the place he worked at. I'll get back to you if I find anything. Out!

Ok, here we go: Norik at Audio Design and Service (North Hollywood). (818) 754-0467

Also, why don't you just take a chance on buying a new screen? Only $200 and readily attainable, as far I can tell...

Last edited by CthonicEwes; 5th February 2015 at 09:27 PM.. Reason: found new info
Old 5th February 2015
  #6
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by plikestechno View Post
I hold out hope that some 3rd party will make new boards one day and all of these Andromedas will come back to life. But this seems to be a common occurence now. Had I known this before I would never have bought one. It would be nice if one of the original engineers could help the cause.
I agree, Plikestechno. I agree 100%. Someone from the inside would need to be the one that could help ressurect the dead Andromedas. The flash chips are out there, now better than ever, and they can easily be programmed with code. The code is the problem, and I'm sure nobody would want to give up their A6 to have their flash chips read with a programmer and risk a failure.

From what I understand, there are two flash chips on the main board. However, I'm not sure if one is the Bootloader and the other is the OS, or if one is the core firmware while the other is the Bootloader and OS. The latter is upgradeable while the other is not. Both cannot be replaced unless they are replaced with working code.

EDIT: Most hardware devices with surface-mount flash chips are programmed via ISP (In-System Programming) these days. So it may be possible to program at least one of the flashes with an ISP (which does indeed exist on the main board).

If it turns out that the my A6's SRAM chips are bad, then I will replace these with modern IC's. I can do that. The cost would be about $12. The battery backup RAM is an obsolete part, but still available from China vendors.

I've successfully repaired an Alesis Ai3 with Chinese-sourced IC's when no other location anywhere in the US had the chips. They were either the real deal or near-perfect counterfeits. Overall, I could not tell the difference, and the chips worked!
Old 5th February 2015
  #7
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by CthonicEwes View Post
Ok, here we go: Norik at Audio Design and Service (North Hollywood). (818) 754-0467

Also, why don't you just take a chance on buying a new screen? Only $200 and readily attainable, as far I can tell...
Thank you! Yes, I may pull the trigger on a new LCD module. Yes, they are pretty much the only parts left. I plan on doing some more investigation and testing and go from there.
Old 6th February 2015
  #8
Gear Addict
 
M.Retra's Avatar
Can anyone tell me if this peculiar resistor exists in their Andy's? This is a 4.7k resistor across what appears to be the capacitor and a pin on one of the SRAM chips (the "Fast SRAM Lower"). This is factory fix, and/or an ECN example I was referring to in my first post. The resistor is not on the schematics.



UPDATE: I carefully peeled away the hot glue. The resistor lead popped off Pin 6. Terrible solder job. A toddler could do better. Anyway, pin 6 is CE (chip enable), and the resistor is "pulling up" this input to a constant high state (at 3.3V), or [EDIT] to provide a pull-up source, and/or provide a stronger current source if the CS output pin on the Coldfire is weak [/EDIT]. Continuity testing reveals that both pin 6's of each SRAM chip are tied together. Therefore, this resistor is pulling up both SRAM chips' CE input. Interesting.

Anyway, I am going to fix this and solder it properly, as it could have been a cold solder joint before.

I also found a broken trace that will need rework on the Front Panel Left PCB. This trace is for an LED, however. Strange as the trace is under the PCB coating. It may have been OK back in the day, but hot/cold cycling worked it away causing quite a gap.

Last edited by M.Retra; 6th February 2015 at 03:36 PM.. Reason: New info. Image link.
Old 6th February 2015
  #9
Gear Addict
 
M.Retra's Avatar
Here's a picture of the broken trace. It's almost like this occurred on the assembly floor before the PCB was coated. Was this not caught by Alesis quality control?

A 10x loupe is your best friend when it comes to SMD work!


Last edited by M.Retra; 6th February 2015 at 03:39 PM.. Reason: Put picture here.
Old 6th February 2015
  #10
Here for the gear
 

Quote:
Originally Posted by plikestechno View Post
Seems like threads like these are coming up weekly now. Andromedas dying everywhere. Not booting or not booting fully. The Flash on the main board is probably dying or dead on all of these but it's not easy to replace.
Are you suggesting every A6 is going to be failing because of issues with flash memory that cannot be replaced or reprogrammed? I sure hope not.

Opal
Old 7th February 2015
  #11
Gear Addict
 
J# Minor's Avatar
 

Quote:
Originally Posted by M.Retra View Post
1. The stupid question: what is the sequence of events that occurs when a normal Andy is powered up?
I don't have a complete picture of this yet, but I'm working on it.

My research suggests (i.e. not definite yet) that the 87C52 paints the splash screen on startup, probably without help from the ColdFire host processor. If you don't see the splash screen, the problem likely is somewhere between the 87C52 and the LCD.

Quote:
Originally Posted by M.Retra View Post
2. Per the Service Manual, the 87C52 processor handles the LCD. So does that mean all of the graphics are handled by the 87C52 too?
Mostly. The host processor talks to the 87C52 via an internal serial line at ~152 Kbps. The protocol consists of short commands, like "clear display" and "draw this text at these coordinates". I haven't discovered yet which chip handles the graphical icons. So far, I haven't found a "draw this bitmap" command, but that doesn't mean it isn't there somewhere. With that said, I suspect it is more of a "draw icon #n at these coordinates" command.

Both the bootloader and OS binaries have a section of Intel HEX format at the end, which probably is intended to be loaded onto the 87C52. I don't know if it is code or data (perhaps bitmap graphics), or when it is loaded, but it is there.

Quote:
Originally Posted by M.Retra View Post
Therefore, how likely is [flash IC replaced and/or reprogrammed] to occur? Have any of you with trouble Andy's had the ?
From wikipedia: Solid state media – such as EPROMs, flash memory and other solid-state drives – stores data using electrical charges, which can slowly leak away due to imperfect insulation. The chip itself is not affected by this, so re-programming it once per decade or so will prevent data decay.
So far, I don't think any user has managed to dump or reprogram the entire A6 flash. It's plausible that it could be done with the right equipment, without removing the flash chip from the board — it's just that no one has tried yet. There are several interested people, so that might change soon.

Quote:
Originally Posted by M.Retra View Post
4. The Service Manual states the SRAM IC's could be problematic.
It says that if the chip an ICSI brand, replace it with ISSI brand.

Quote:
Originally Posted by M.Retra View Post
what ECN's are out there?
If anyone knows, they aren't talking.
Old 7th February 2015
  #12
Jesus! Reading this thread makes me want to sell all my gear while it's working and go 100% ITB.
Old 7th February 2015
  #13
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by J# Minor View Post
I don't have a complete picture of this yet, but I'm working on it.

My research suggests (i.e. not definite yet) that the 87C52 paints the splash screen on startup, probably without help from the ColdFire host processor. If you don't see the splash screen, the problem likely is somewhere between the 87C52 and the LCD.
If this is correct, this is quite interesting as my experience in programming 8051-style processors was somewhat limited to character displays, not graphical displays. But as you say below, the LCD is still setup in a grid format, just more points to be programmed rather than alphanumeric characters. Therefore, I suppose it is indeed possible for the 87C52 to handle all graphics.

In order to prove that the 87C52 can do at least the splash screen, it may come to the point to remove the main ribbon cable, and power the Front Panel Left PCB using a desktop power supply and see what happens.

The startup sequence I was actually looking for was much more simple--more external behavior rather than internal. I was merely interested, as a normal user would experience, in what happens when the main power switch is turned on. In other words, the splash screen pops up, go away, and the lights come on. But getting more in depth with the internal events would certainly be more useful. Because I can't see anything on the LCD, I am not 100% sure of what's really happening. So the goal was to confirm that I didn't need to press a soft button or other key to continue, so to speak.

Quote:
Originally Posted by J# Minor View Post
Mostly. The host processor talks to the 87C52 via an internal serial line at ~152 Kbps. The protocol consists of short commands, like "clear display" and "draw this text at these coordinates". I haven't discovered yet which chip handles the graphical icons. So far, I haven't found a "draw this bitmap" command, but that doesn't mean it isn't there somewhere. With that said, I suspect it is more of a "draw icon #n at these coordinates" command.
I assume this serial connection using the RXD and TXD ports? I will need to check those ports for action.

Quote:
Originally Posted by J# Minor View Post
Both the bootloader and OS binaries have a section of Intel HEX format at the end, which probably is intended to be loaded onto the 87C52. I don't know if it is code or data (perhaps bitmap graphics), or when it is loaded, but it is there.
Perhaps not. The 87C52 used in the A6 (P87C52UBBB) is a one-time-programmable device. Any internal flash (8K) would be programmed once, most likely from the factory, if needed. But as far as I can tell at this point, the 87C52 is "forced" to read "external", as resistor R42 exists and it is tying EA (external access enable) low. This permanently has the 87C52 read the EPROM chip for any code and load it into internal RAM (256x8) for use.

And the EPROM chip cannot be reprogrammed without being erased with UV light. These things are solid. The reason why Alesis probably chose a 1986-vintage UV-erasable EPROM chip was to eliminate the risk of an erasure.

Pulling the code off the EPROM would be easy and guarantee life after death for the Andromeda, at least for the Front Panel Left board. Any chip programmer could do it. The EPROM is easy. On the other hand, the SMD flash chip is the problem.

On a side note, I love 8051 microcontrollers! I haven't programmed one for quite some time, but I am quite familiar with the hardware and internal functions.

Quote:
Originally Posted by J# Minor View Post
From wikipedia: Solid state media – such as EPROMs, flash memory and other solid-state drives – stores data using electrical charges, which can slowly leak away due to imperfect insulation. The chip itself is not affected by this, so re-programming it once per decade or so will prevent data decay.
So far, I don't think any user has managed to dump or reprogram the entire A6 flash. It's plausible that it could be done with the right equipment, without removing the flash chip from the board — it's just that no one has tried yet. There are several interested people, so that might change soon.
Yeah, I suppose any programmable device can lose it's programming over time. I would think, however, that it is quite rare. However, I always thought that EPROM's and other "flash" based devices are physically altered when programmed. Take for example the UV-erasable EPROM device. That's no electrical charge; the UV light is physically "resetting" the silicon transistors.

I hope that things turn around for the Andromeda. I am willing to contribute to the cause. I think it's absolutely ridiculous that a behemoth synthesizer is put out to pasture and left to die.

Quote:
Originally Posted by J# Minor View Post
It says that if the chip an ICSI brand, replace it with ISSI brand.
I tried looking up information on ICSI brand chips, and couldn't find anything. Something tells me that ISSI and ICSI is one in the same, like ISSI bought out ICSI a long time ago and nobody cared to notice. The chips on my main board, revision 'B', appear to be ICSI.

Quote:
Originally Posted by J# Minor View Post
If anyone knows, they aren't talking.
This is a big part of the problem. The A6 doesn't have a snowball's chance in Hell unless we all compile data and get working on a solution. Hoarding parts to the point of unavailability is not the solution. It would be wonderful if Alesis/Numark realized the "cult status" of this machine and contributed as well, supplying HEX code for reprogramming future flash ROMS and EPROMS.

Overall, I sense there is a certain fear of the Andromeda--a fear of the inevitable and a fear of disturbing the sleeping beast. Is that the reason for the "cult status?" Is the cult status to worship the Andromeda while it dies a slow death?

J# Minor, I appreciate your contribution to this thread. Thank you! I will be taking knowledge from your post and putting it towards my troubleshooting efforts.

Quote:
Originally Posted by plaid_emu View Post
Jesus! Reading this thread makes me want to sell all my gear while it's working and go 100% ITB.
Now let's not get foolish!

Last edited by M.Retra; 7th February 2015 at 06:20 PM.. Reason: More information...
Old 7th February 2015
  #14
I really think you should get in touch with the other guy on here who is looking at the OS. Between the two of you, I think you would have a greater chance of success.

Keep up the great work!
Old 7th February 2015
  #15
Lives for gear
 
NEXUS-6's Avatar
 

Quick while you still can everybody sell your Andromeda its a disaster waiting to happen & will explode !!
Old 7th February 2015
  #16
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by kazper View Post
I really think you should get in touch with the other guy on here who is looking at the OS. Between the two of you, I think you would have a greater chance of success.

Keep up the great work!
Thanks, and I will see if I can locate him!

Quote:
Originally Posted by NEXUS-6 View Post
Quick while you still can everybody sell your Andromeda its a disaster waiting to happen & will explode !!
Or collect them all and bury them in the desert!
Old 8th February 2015
  #17
Gear Addict
 
J# Minor's Avatar
 

Quote:
Originally Posted by M.Retra View Post
Thanks, and I will see if I can locate him!
That's me.
Old 8th February 2015
  #18
Gear Addict
 
M.Retra's Avatar
Major, major, MAJOR breakthrough guys!

Old 8th February 2015
  #19
Gear Addict
 
J# Minor's Avatar
 

^^^ Signal acquired. ^^^

So this is the front panel left PCB by itself? Sweet!

Quote:
Originally Posted by M.Retra View Post
I assume this serial connection using the RXD and TXD ports? I will need to check those ports for action.
Yes. 156250 baud, no parity, 8 data bits, 1 stop bit.

If it helps, it is pins 12 and 13 of the ribbon cable between the main board and front panel left board. The corresponding pins on the ColdFire are TXD0 and RXD0.

Quote:
Originally Posted by M.Retra View Post
The 87C52 used in the A6 (P87C52UBBB) is a one-time-programmable device. Any internal flash (8K) would be programmed once. ... And the EPROM chip cannot be reprogrammed without being erased with UV light.
There's also a 128KB SRAM chip specifically for the 87C52, so whatever payload the host processor sends probably ends up there.

Quote:
Originally Posted by M.Retra View Post
the SMD flash chip is the problem.
With luck, there's a device that can be connected to the debug port on the main PCB in order to read/write the flash directly, among other things.

Quote:
Originally Posted by M.Retra View Post
I always thought that EPROM's and other "flash" based devices are physically altered when programmed.
Just a hypothesis based on a few things I read. Would be interesting if we could resurrect a dead A6 by rewriting the flash, though.
Old 8th February 2015
  #20
Lives for gear
 
CthonicEwes's Avatar
Well, I guess your LCD is ok. How did you fix it? Did you just use the LH PCB as J#minor notes above? Congrats!

How about the rest of the synth? I hate to say this, but if you can't get past the slash screen you may still be in deep sh*t.

Anyway, check the ribbon connection to the modwheel and jiggle the heck out of the modwheel. I have read about modwheel jitter problems before. Sadly, I would occasionally get that screen when I tried to get my A6 going. Repair was beyond my abilities, but you seem to be very smart. Please keep us updated. I've got my fingers crossed for you!
Old 9th February 2015
  #21
Gear Addict
 
M.Retra's Avatar
OK, I'm back! Sorry for the delayed response! As you can see from my post above, the LCD screen is working! When I first saw the splash screen pop up, I nearly fell out of my chair!

And as you can see below, the software is running and getting past the splash screen!



I apologize for the crappy images above and below. I was so excited that I decided to make a video. But then after listening to myself, I decided I don't do very good voice-overs, so I axed the project! Anyway, the video capture below shows my test arrangement.



Overall, what we're looking at here is an old Wavetek 182A function generator, which I had to get working after years of storage, generating and supplying a 3.0MHz square wave (albeit a distorted square wave). This square wave is fed into the LCD module. Here's why:

While troubleshooting and looking for data streams on the ribbon cable, I came across an important discovery. On the back of the LCD module is a Toshiba T6963C LCD driver IC. This IC handles all of the data communication between the LCD itself and the 87C52 microcontroller. I looked up the datasheet for the chip, and I learned that the T6963C has an internal oscillator. This piqued my interest, so using the pinout diagram in the datasheet, I put my oscilloscope probe on the X0 pin, which is supposed to be the clock/oscillator output. I wasn't getting anything. I then moved the probe over to the X1 pin (clock input). Again, nothing. So either the T6963C was defective, or the clock is coming from somewhere else.

Enter the 3.0MHz ceramic resonator.



That little guy was found to be dead! Basically, a ceramic resonator functions the same as a crystal oscillator, but it is a cheaper and less precise. The ceramic resonator was here on the LCD module board:



Why 3.0MHz? According to the T6963C datasheet, the clock frequency is determined by the number of rows and columns in the LCD screen. The more rows and columns to populate with a "pixel", then the faster the clock frequency. Think of the clock signal as the refresh rate, like in TV's and LCD monitors for your computer.

So I had an idea--what if I unsoldered and removed the bad resonator and hooked up an external clock source to the X1 pin of the T6963C? Well, you guess it...the LCD module came to life! What's interesting to note is the fact that my Wavetek function generator is completely unstable and nowhere near being dead-on accurate like a crystal oscillator. Well, I guess the T6963C doesn't care! It just needs a clock!

So the fix now will to order a 3.00MHz ceramic resonator. I will also order a 3.00MHz crystal oscillator and install it instead, as I prefer them more, but I will need to swap out SMD capacitors also in order to provide stability.

Now here's the kicker--without the LCD clock source, the Andy will not, I repeat WILL NOT boot. This...is...huge, IMO, and it could be an explanation as to why many Andromedas out there do not get past the splash screen! Perhaps once a big graphic is displayed on the screen, like the splash screen, then the T6963C craps out (and hangs), perhaps due to a bad/weak clock signal.

Certainly the above, last statement is plausible. I find it truly unbelievable that the Andy will not boot with this clock signal removed. This contradicts typical LCD module operation and the Andy schematics. By looking at the Service Manual schematics, there is no feedback path from the LCD back to the 87C52. So is there bidirectional data going on instead? Maybe. Further investigation may be warranted.

Apparently, a bad ceramic resonator on the LCD module can stop an Andromeda dead in its tracks.

The above sentence is worth reading again.

So there you have it. More than you expected, I'm sure. I am very happy now, but not getting my hopes up 100%. I still need go back and:
  • Replace the bad ceramic resonator with a crystal oscillator for more reliable operation. COMPLETE. SEE POST #35.
  • Fix the pull-up resistor on the SRAM chip to be more permanent. COMPLETE. SEE POST #49.
  • Go thru ASIC osc/filter tuning procedures. COMPLETE. SEE POST #27.
  • Replace all of the electrolytic caps on the ASIC and Master Analog board. COMPLETE. ONLY REPLACED CAPS ANALOG MASTER BOARD. SEE POST #50.
  • Add a static-grounding wire to the LCD module if none of the screw posts tie the board to the digital-ground plane/pin/wire. NOT NEEDED; SCREWS TIGHTEN DOWN ON GROUND PLANE.
  • Redo tuning procedures, as needed, after capacitor replacements. NOT NEEDED
  • Clean and lubricate all pots. DECIDED NOT TO. MORE ON THIS LATER...
  • Fix a broken key. COMPLETE. SIMPLE FIX; SEE SERVICE MANUAL.
  • Tighten the right end cap/cheek. COMPLETE.
  • Replace memory backup battery. COMPLETE. SEE POST #24.

Last edited by M.Retra; 10th February 2015 at 01:11 PM.. Reason: Added: "and hangs..."; New Edit: Should be "T6963C".
Old 9th February 2015
  #22
Lives for gear
 
CthonicEwes's Avatar
Outstanding! Have you gotten audio out of it yet? Did you run some of the push button diagnostics? I'm not sure this is the solution to all the woes people had, since the problem wasn't a blank screen, but rather a hanging screen.
Old 9th February 2015
  #23
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by CthonicEwes View Post
Outstanding! Have you gotten audio out of it yet? Did you run some of the push button diagnostics? I'm not sure this is the solution to all the woes people had, since the problem wasn't a blank screen, but rather a hanging screen.
Yes sir! I have indeed got some audio out of it, although it was unusually low pitch...like crazy low pitch! So maybe some oscillator setup? Not quite sure. Above all else, at this time, it does respond to the keyboard. I haven't had time to investigate further. I've lost enough sleep and time over this thing, I need to get back to reality for a bit! Haha!

The bad resonator may not be the solution to all Andy problems, especially the stuck splash screen, but it is certainly plausible. Therefore, I have revised my previous post just a tad.
Old 9th February 2015
  #24
Gear Addict
 
M.Retra's Avatar
Memory-Backup Battery, Battery Holder Replacement

Whew! I forgot to mention that I also replaced the soldered, memory-backup battery with a new battery and holder that allows for easy changing.

The battery is a CR2032, and the battery holder is from Radio Shack, Catalog No. 2700009. Fresh battery voltage is about 3.2V.



For a similar fix, I should mention that a guy over on Matrixsynth.com put a jumper between the two, positive terminal vias (holes in printed circuit boards). The main board revision in my A6 is 'B', and it was not necessary to put a jumper between the two vias for the positive terminal. On the revision 'B' board, a trace exists and connects both positive-terminal vias together. I cannot speak for other board revisions. YMMV.

Note: if your Andy does a lot of traveling, then it would probably be a good idea to place a dab of hot glue between the battery and holder.
Old 9th February 2015
  #25
Lives for gear
 
CthonicEwes's Avatar
Did you do the tuning sequence for the VCOs and VCAs? Did you test for whether the voices pass/fail? These are obvious things to do before doing anything else. Of course, you probably already did all this. Just throwing out the basics, as I'm not a guru!
Old 9th February 2015
  #26
Gear Addict
 
J# Minor's Avatar
 

Quote:
Originally Posted by M.Retra View Post
I find it truly unbelievable that the Andy will not boot with this clock signal removed. This contradicts typical LCD module operation and the Andy schematics. By looking at the Service Manual schematics, there is no feedback path from the LCD back to the 87C52. So is there bidirectional data going on instead?
Yes, the communication is bidirectional, across an 8-bit data bus. The 87C52 can write commands/data to, and can read status/data from, the T6963C. This is all spelled out quite clearly in the datasheet (mine is for T6963CFG, specifically).

In fact, the datasheet says (page 5):
Quote:
The status must be checked before data or commands are sent.
There also are flowcharts on page 10. If the 87C52 is following this procedure, it will notice a problem and not send the command to the T6963C. Perhaps the 87C52 halts in this case.

Now, the 87C52 is the front panel controller (FPC). In addition to the LCD, it also monitors the buttons, I believe. During boot, the A6 host processor has to communicate with the FPC to detect if any soft buttons are depressed, as the soft buttons are how you get to the maintenance modes (OS update, etc.). If the FPC does not respond (having halted after a LCD error), it is plausible that the host would refuse to continue with the boot process. After all, if the host can't talk to the front panel, there's not much synthesis that can be done.
Old 10th February 2015
  #27
Gear Addict
 
M.Retra's Avatar
Quote:
Originally Posted by CthonicEwes View Post
Did you do the tuning sequence for the VCOs and VCAs? Did you test for whether the voices pass/fail? These are obvious things to do before doing anything else. Of course, you probably already did all this. Just throwing out the basics, as I'm not a guru!
OK, here's the report:



Other tests:

MIDI Test: Passed
LED Test: Passed, all lit--thanks to that trace fix!
LCD Test: Passed, solid black
SSRAM Test: Passed

Sound Check: I hear sound when keys are played!

I do have some glitchy pots, so it makes the screen jump around, but that fix/cleaning is coming soon.

OS 1.40.12
Bootloader 1.20.37


It appears that the A6 is up-to-date. OS 1.40.12 is the latest, and last, update for the A6.
Old 10th February 2015
  #28
Lives for gear
 
CthonicEwes's Avatar
Yay! Seems like you're nearly there. Very impressive!
Old 10th February 2015
  #29
Gear Addict
 
J# Minor's Avatar
 

So cool to see an Andy back from the dead!
Old 11th February 2015
  #30
Lives for gear
 
CthonicEwes's Avatar
Waiting on pins and needles for your next update.

M. Retra and J# Minor: I'm guessing you guys are computer engineers. I'm astounded by the amount of knowledge you both have. Is music electronics your hobby, or do you work on music products for a living? Any chance you guys want to team up and design a new synth? I'd be really curious to see/hear what you could come up with. Seriously!
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump