Emu XL-7 & MP-7 User's Group

Does anyone have a solution for an FX RAM failed error?

I have a backup P2500 that hasn't been used in several years. It used to work flawlessley. However, when I went to test it, there was an awful noise coming through the headphones. Everything testable in diagnostics passes, except FX RAM.  I cannot imagine it would would have failed from lack of being used, but who knows. Any thoughts, suggestions? Is it possible to replace the FX RAM? If so, how? Thanks.

--Steve

Footnotes: 

(1) Since only the main outs are connected to the FX system, I suppose I could always use it was a non-FX E-Mu/sequencer/controller using the sub outs. Still, it would be nice to have the FX working.

(2) Putting FX into bypass mode does not get rid of the noise.
--- In xl7@yahoogroups.com, "steve_the_composer" <smw-mail@...> wrote:
>
> Does anyone have a solution for an FX RAM failed error?
> 
> I have a backup P2500 that hasn't been used in several years. It used to work flawlessley. However, when I went to test it, there was an awful noise coming through the headphones. Everything testable in diagnostics passes, except FX RAM.  I cannot imagine it would would have failed from lack of being used, but who knows. Any thoughts, suggestions? Is it possible to replace the FX RAM? If so, how? Thanks.
> 
> --Steve
>

Hey steve.

If the P2500 mother board is identical to an XL7 motherboard (I suspect so, but...) FX RAM is the SOJ28 (the one with missing central pins) package to the front left of the motherboard - U35; right next to the ROM slots. This is not memory used by the microcontroller (which is towards the right and back on the motherboard in a TSOP 54 package).

The FX RAM (81C4256 = 256kBit) is used to produce a pipeline by the large IC next to it "IC405" (U34) for sound effects (not the FX bus) so that you'll probably find that any sound that has any effect (eg chorus) will be bad.

It is unusual that a DRAM would fail, but if you have access to a hot air solder rework station you could try reworking the solder joints (which may have become dry). You also need to consider the bypass capacitors next to it and solder joints on U34 (with which it communicates). If the DRAM has not failed (and its not just a dry joint or two) then U34 could be failing which means that you're unlikely to ever get good sound out of the thing wihtout stipping all effects [making the sound output very bland].

If you are not competent to do the rework yourself, then you probably need to find someone who is to avoid damage to your unit [not something you should attack if the soldering iron is not your friend].

Unfortuantely 81C256's are not common at all these days, and the package makes removal a challenge without a solder rework station [and then its trivial]. So if you wanted to try replacing it...


 



________________________________
From: steve_the_composer <smw-mail@...>
To: xl7@yahoogroups.com
Sent: Tue, March 16, 2010 7:31:21 PM
Subject: [xl7] Re: HELP: FX RAM failed error

  
Footnotes: 

(1) Since only the main outs are connected to the FX system, I suppose I could always use it was a non-FX E-Mu/sequencer/ controller using the sub outs. Still, it would be nice to have the FX working.

(2) Putting FX into bypass mode does not get rid of the noise.
--- In xl7@yahoogroups. com, "steve_the_composer " <smw-mail@.. .> wrote:
>
> Does anyone have a solution for an FX RAM failed error?
> 
> I have a backup P2500 that hasn't been used in several years. It used to work flawlessley. However, when I went to test it, there was an awful noise coming through the headphones. Everything testable in diagnostics passes, except FX RAM. I cannot imagine it would would have failed from lack of being used, but who knows. Any thoughts, suggestions? Is it possible to replace the FX RAM? If so, how? Thanks.
> 
> --Steve
>

From: Jack Pratt
Subject: Re: [xl7] Re: HELP: FX RAM failed error
To: xl7@yahoogroups.com
Date: Tuesday, 16 March, 2010, 20:15

Hey steve.

If the P2500 mother board is identical to an XL7 motherboard (I suspect so, but...) FX RAM is the SOJ28 (the one with missing central pins) package to the front left of the motherboard - U35; right next to the ROM slots. This is not memory used by the microcontroller (which is towards the right and back on the motherboard in a TSOP 54 package).

The FX RAM (81C4256 = 256kBit) is used to produce a pipeline by the large IC next to it "IC405" (U34) for sound effects (not the FX bus) so that you'll probably find that any sound that has any effect (eg chorus) will be bad.

It is unusual that a DRAM would fail, but if you have access to a hot air solder rework station you could try reworking the solder joints (which may have become dry). You also need to consider the bypass capacitors next to it and solder joints on U34 (with which it communicates). If the DRAM has not failed (and its not just a dry joint or two) then U34 could be failing which means that you're unlikely to ever get good sound out of the thing wihtout stipping all effects [making the sound output very bland].

If you are not competent to do the rework yourself, then you probably need to find someone who is to avoid damage to your unit [not something you should attack if the soldering iron is not your friend].

Unfortuantely 81C256's are not common at all these days, and the package makes removal a challenge without a solder rework station [and then its trivial]. So if you wanted to try replacing it...

From: steve_the_composer
To: xl7@yahoogroups.com
Sent: Tue, March 16, 2010 7:31:21 PM
Subject: [xl7] Re: HELP: FX RAM failed error

Footnotes:

(1) Since only the main outs are connected to the FX system, I suppose I could always use it was a non-FX E-Mu/sequencer/ controller using the sub outs. Still, it would be nice to have the FX working.

(2) Putting FX into bypass mode does not get rid of the noise.
--- In xl7@yahoogroups. com, "steve_the_composer " wrote:
>
> Does anyone have a solution for an FX RAM failed error?
>
> I have a backup P2500 that hasn't been used in several years. It used to work flawlessley. However, when I went to test it, there was an awful noise coming through the headphones. Everything testable in diagnostics passes, except FX RAM. I cannot imagine it would would have failed from lack of being used, but who knows. Any thoughts, suggestions? Is it possible to replace the FX RAM? If so, how? Thanks.
>
> --Steve
>

It depends on the implementation as to how the 'frequency blurring' of the chorus is done (and there are a wide variety of choruses - a simple delay line can give a chorus of sorts). That may or may not involve the pipeline, but something like a reverb certainly does (this is just 'frequency blurring' with a low pass filter). Since I don't have any information on the 'IC405' part from EMU I'm just indicating the type of effect that may be subject to pipeline...


________________________________
From: massimo <massenmedium@...>
To: xl7@yahoogroups.com
Sent: Wed, March 17, 2010 6:22:19 AM
Subject: Re: [xl7] Re: HELP: FX RAM failed error

  
Just an aside but isn't the chorus done with doubling and detuning?

There's two different choruses - one that's done at the voice level (by detuning an additional voice - which is why it eats polyphony) and one that's done at the effect level (by using the tank RAM as Jack mentions).

The FX RAM will mess with pretty much any signal coming through the box.  The G05 chip uses this memory for just about everything.  It may even use it for routing signals.  Did you try the subs?  I would imagine that even though they're dry, they might still be affected by the bad RAM.  Also note that it might be the interface to RAM that's suspect, not just the chip itself.  You could go physically pushing chips on the board to see if you hear anything change.  It's actually kinda fun to debug audio circuits...

-Aaron


----- Original Message ----
From: Jack Pratt <woodsworth1@...>
To: xl7@yahoogroups.com
Sent: Tue, March 16, 2010 3:36:28 PM
Subject: Re: [xl7] Re: HELP: FX RAM failed error



It depends on the implementation as to how the 'frequency blurring' of the chorus is done (and there are a wide variety of choruses - a simple delay line can give a chorus of sorts). That may or may not involve the pipeline, but something like a reverb certainly does (this is just 'frequency blurring' with a low pass filter). Since I don't have any information on the 'IC405' part from EMU I'm just indicating the type of effect that may be subject to pipeline...


________________________________
From: massimo <massenmedium@...>
To: xl7@yahoogroups.com
Sent: Wed, March 17, 2010 6:22:19 AM
Subject: Re: [xl7] Re: HELP: FX RAM failed error

  
Just an aside but isn't the chorus done with doubling and detuning?


      


------------------------------------

Yahoo! Groups Links

Thanks for the details, Jack. Such repairs are definitely beyond my skill set. I might be able to find someone else who can do it. I will look at the motherboard for the chips to see what I can find. 

--Steve



--- In xl7@yahoogroups.com, Jack Pratt <woodsworth1@...> wrote:
>
> Hey steve.
> 
> If the P2500 mother board is identical to an XL7 motherboard (I suspect so, but...) FX RAM is the SOJ28 (the one with missing central pins) package to the front left of the motherboard - U35; right next to the ROM slots. This is not memory used by the microcontroller (which is towards the right and back on the motherboard in a TSOP 54 package).
> 
> The FX RAM (81C4256 = 256kBit) is used to produce a pipeline by the large IC next to it "IC405" (U34) for sound effects (not the FX bus) so that you'll probably find that any sound that has any effect (eg chorus) will be bad.
> 
> It is unusual that a DRAM would fail, but if you have access to a hot air solder rework station you could try reworking the solder joints (which may have become dry). You also need to consider the bypass capacitors next to it and solder joints on U34 (with which it communicates). If the DRAM has not failed (and its not just a dry joint or two) then U34 could be failing which means that you're unlikely to ever get good sound out of the thing wihtout stipping all effects [making the sound output very bland].
> 
> If you are not competent to do the rework yourself, then you probably need to find someone who is to avoid damage to your unit [not something you should attack if the soldering iron is not your friend].
> 
> Unfortuantely 81C256's are not common at all these days, and the package makes removal a challenge without a solder rework station [and then its trivial]. So if you wanted to try replacing it...
> 
> 
>  
> 
> 
> 
> ________________________________
> From: steve_the_composer <smw-mail@...>
> To: xl7@yahoogroups.com
> Sent: Tue, March 16, 2010 7:31:21 PM
> Subject: [xl7] Re: HELP: FX RAM failed error
> 
>   
> Footnotes: 
> 
> (1) Since only the main outs are connected to the FX system, I suppose I could always use it was a non-FX E-Mu/sequencer/ controller using the sub outs. Still, it would be nice to have the FX working.
> 
> (2) Putting FX into bypass mode does not get rid of the noise.
> --- In xl7@yahoogroups. com, "steve_the_composer " <smw-mail@ .> wrote:
> >
> > Does anyone have a solution for an FX RAM failed error?
> > 
> > I have a backup P2500 that hasn't been used in several years. It used to work flawlessley. However, when I went to test it, there was an awful noise coming through the headphones. Everything testable in diagnostics passes, except FX RAM. I cannot imagine it would would have failed from lack of being used, but who knows. Any thoughts, suggestions? Is it possible to replace the FX RAM? If so, how? Thanks.
> > 
> > --Steve
> >
>

not quite sure where this message went - it's not in pending...


----- Forwarded Message ----
From: Aaron Eppolito <synthesis77@...>
To: xl7@yahoogroups.com
Sent: Tue, March 16, 2010 7:37:38 PM
Subject: Re: [xl7] Re: HELP: FX RAM failed error

There's two different choruses - one that's done at the voice level (by detuning an additional voice - which is why it eats polyphony) and one that's done at the effect level (by using the tank RAM as Jack mentions).

The FX RAM will mess with pretty much any signal coming through the box.  The G05 chip uses this memory for just about everything.  It may even use it for routing signals.  Did you try the subs?  I would imagine that even though they're dry, they might still be affected by the bad RAM.  Also note that it might be the interface to RAM that's suspect, not just the chip itself.  You could go physically pushing chips on the board to see if you hear anything change.  It's actually kinda fun to debug audio circuits...

-Aaron


----- Original Message ----
From: Jack Pratt <woodsworth1@...>
To: xl7@yahoogroups.com
Sent: Tue, March 16, 2010 3:36:28 PM
Subject: Re: [xl7] Re: HELP: FX RAM failed error



It depends on the implementation as to how the 'frequency blurring' of the chorus is done (and there are a wide variety of choruses - a simple delay line can give a chorus of sorts). That may or may not involve the pipeline, but something like a reverb certainly does (this is just 'frequency blurring' with a low pass filter). Since I don't have any information on the 'IC405' part from EMU I'm just indicating the type of effect that may be subject to pipeline...


________________________________
From: massimo <massenmedium@...>
To: xl7@yahoogroups.com
Sent: Wed, March 17, 2010 6:22:19 AM
Subject: Re: [xl7] Re: HELP: FX RAM failed error

  
Just an aside but isn't the chorus done with doubling and detuning?


      


------------------------------------

Yahoo! Groups Links

Just incidentally, the same basic circuit is used for the sound generation in all P2K series units and command stations [The command stations have a different micro & the FLASH/SDRAM chips (it has an SDRAM chip instead of a DRAM SIMM) are different] So if you have a P2K module that's dead it could be the source of spare parts. I noticed that on a P2K the FX RAM is right next to the micro [U29] with 'IC405' beside it.

The FX RAM will only have an interface to the FX chip  ['IC405'] next to it, so only the DRAM, that chip and the nearby passives are going to be contributing to a FX RAM error.

I'd look at it for you, but I doubt that its worth the trip to Australia...




________________________________
From: steve_the_composer <smw-mail@...>
To: xl7@yahoogroups.com
Sent: Wed, March 17, 2010 8:15:11 PM
Subject: [xl7] Re: HELP: FX RAM failed error

  
Thanks for the details, Jack. Such repairs are definitely beyond my skill set. I might be able to find someone else who can do it. I will look at the motherboard for the chips to see what I can find. 

--Steve

--- In xl7@yahoogroups. com, Jack Pratt <woodsworth1@ ...> wrote:
>
> Hey steve.
> 
> If the P2500 mother board is identical to an XL7 motherboard (I suspect so, but...) FX RAM is the SOJ28 (the one with missing central pins) package to the front left of the motherboard - U35; right next to the ROM slots. This is not memory used by the microcontroller (which is towards the right and back on the motherboard in a TSOP 54 package).
> 
> The FX RAM (81C4256 = 256kBit) is used to produce a pipeline by the large IC next to it "IC405" (U34) for sound effects (not the FX bus) so that you'll probably find that any sound that has any effect (eg chorus) will be bad.
> 
> It is unusual that a DRAM would fail, but if you have access to a hot air solder rework station you could try reworking the solder joints (which may have become dry). You also need to consider the bypass capacitors next to it and solder joints on U34 (with which it communicates) . If the DRAM has not failed (and its not just a dry joint or two) then U34 could be failing which means that you're unlikely to ever get good sound out of the thing wihtout stipping all effects [making the sound output very bland].
> 
> If you are not competent to do the rework yourself, then you probably need to find someone who is to avoid damage to your unit [not something you should attack if the soldering iron is not your friend].
> 
> Unfortuantely 81C256's are not common at all these days, and the package makes removal a challenge without a solder rework station [and then its trivial]. So if you wanted to try replacing it...
> 
> 
>  
> 
> 
> 
> ____________ _________ _________ __
> From: steve_the_composer <smw-mail@.. .>
> To: xl7@yahoogroups. com
> Sent: Tue, March 16, 2010 7:31:21 PM
> Subject: [xl7] Re: HELP: FX RAM failed error
> 
>   
> Footnotes: 
> 
> (1) Since only the main outs are connected to the FX system, I suppose I could always use it was a non-FX E-Mu/sequencer/ controller using the sub outs. Still, it would be nice to have the FX working.
> 
> (2) Putting FX into bypass mode does not get rid of the noise.
> --- In xl7@yahoogroups. com, "steve_the_composer " <smw-mail@ .> wrote:
> >
> > Does anyone have a solution for an FX RAM failed error?
> > 
> > I have a backup P2500 that hasn't been used in several years. It used to work flawlessley. However, when I went to test it, there was an awful noise coming through the headphones. Everything testable in diagnostics passes, except FX RAM. I cannot imagine it would would have failed from lack of being used, but who knows. Any thoughts, suggestions? Is it possible to replace the FX RAM? If so, how? Thanks.
> > 
> > --Steve
> >
>

I think you deserve some encouragement !

Keep up the good work :)

We'll wait it doesn't matter.



--- In xl7@yahoogroups.com, Jack Pratt <woodsworth1@...> wrote:
>
> Well I have some good & some bad news...
> 
> Good news first. 
> 
> I did the schematic and the PCB layout for the base (programmer) board 
> [attached]. This board allows the user to read and write ROM & FLASH SIMMS 
> including the ones that I'm making; and program the FLASH SIMMs and the new ones 
> that I'm making. It also allows the programming of the CPLD and micro on the 
> SIMM [through the second SIMM socket with the SIMM inserted upside down]. This 
> was an easier work than the SIMM so I was able to complete it with 8 mil track 
> and space. That means it will be relatively inexpensive to make (compared to the 
> SIMM) but will be more expensive than the SIMM because of the size [5.4" x 3.8" 
> cf 4.25" x 1.1"]. I will also save some $$ by not having the bottom screen 
> printed. Components on the bottom layer are some 100uF caps and some 10K 
> resistor packs (and a single zero ohm link) - but there's nothing on the bottom 
> layer that needs to be identified by the screen print.
> 
> I need to make at least one to read my ROM SIMMs to determine the format, and to 
> develop the software for the CPLD and micro. No one else really needs one but if 
> anyone's interested in getting one. The cost would be around USD200. Please note 
> that although these boards can be used to (re)program the CPLD and micro, once 
> the SIMM has the mini-USB connector attached it won't fit into the SIMM socket. 
> So if you want one of these please give me some indication. Otherwise I'll just 
> make a couple for my own use.
> 
> Just on that, I have 22 of the 23 ROM SIMMs that EMU produced. I don't have a 
> copy of the Composer ROM from a proteus 2500. If anyone has such a beast and is 
> willing to part with it, then I would be happy to pay a reasonable price for it 
> or come to some other arrangement. It would have a sticker on it with the number 
> AP<xxxxx>-20 where <xxxxx> is one of 638, 10172, 10498, or 12068 [probably the 
> latter one]
> 
> Now the bad news.
> 
> I thought that I had some SAM7SE micros that I purcahsed earlier. However, I 
> can't find any. So I looked online and the parts seem to be on long lead time 
> allocation [around four months]. I will order some ASAP and might be able to get 
> hold of some samples, but it will probably four or five months before I can make 
> quantities for sale. :(
> 
> Oh well. we do what we can...
>

>>I think you deserve some encouragement !

Keep up the good work :)
 
We'll wait it doesn't matter.<<

agreed. :-)

duncan.