Old Ibanez Midi Loop Switcher Slow to Switch

[Rane]

Hi, I have a question about a midi switcher that I own.  It cuts out just for a second when it switches.  The unit was manufactured somewhere between 1986 and 1988.  I just wondered if anyone could tell me if this is just a normal thing to expect from midi equipment this old. 

One thing I did notice is what I think is a battery.  Picture below.  I doubt a battery from the 80's doesn't need replacing.  Could this be a problem?  Honestly, I'm clueless.




Here are some more pictures.  I can't find a schematic, and I realize pictures are probably not that helpful when it comes to insight on how the thing works.












Anyway, if someone could help, I'd really appreciate it.  I planned to buy a more modern switcher, but if I could get this one to work or even if I could just know for sure if it's broken or not, I'd be really happy.

Thanks!

[iaresee]

The battery is most likely there to provide power to the EEPROMs that store your patches and system settings (like which channel to listen to, etc.). So when you power down the unit your patch and system settings aren't lost. If you aren't losing those settings between power down/up cycles you don't need to replace the battery.

Does the unit use mechanical relay switched to move things in and out of the signal path? If so: is the delay uniform on all the loops? If it's non-uniform I'd suspect some relays are just getting old and need replacing. They're mechanical and can wear out, so not unexpected.

If the unit doesn't use relays for switching, or if the delay is uniform, it might be harder to figure out what's going wrong. Do you know if the power supply for the unit is okay? Is everything getting the current it needs to do its switching? If there's a ramp up/down type thing happening on the power rails when you switch something in and out, maybe you've got power supply components like the caps that remove the AC ripple starting to die?

Also consider what you're using to send the MIDI signals to the unit -- this unit may not handle the latest MIDI specs. Consider isolating it with whatever you're using to send the MIDI PC/CC signals with. Make sure you're transmitting *only* the MIDI signals the unit responds to. If it responds to PC change numbers make sure your unit is bombarding it with a bunch of CC state information as well as the PC change signal when you change patches. Could be you're bogging down the processor dealing with extra messages it might just be throwing away.

[Rane]

Thanks so much for your reply.  So the battery's not the problem then.  I think I've heard of that before... about volatile and non-volatile memory.

Unfortunately, I don't think this thing uses relays at all.  There are two, but I think they're only used for switching inputs and outputs, not loops. ...  At least, that's what I gather from this diagram I found:

http://www.ibanezcollectors.com/gallery/displayimage.php?pid=408&fullsize=1

Also, I don't think it's the midi controller as mine only sends PC messages and I only have one or two channels sending messages at all.  Also, the thing switches at the same speed when using either the midi controller or the buttons on the front.

I apologize, I'm not sure what you mean about the power rails.  The power supply caps are the ones that are on the circuit board with the fuse, correct?  Would you suggest replacing these?

It only cuts out for a split second between the switch.  Do you think it's the way thing was designed?  Maybe just the limitations of old technology?   

[iaresee]

"Digital Controller Patching Programmer" doesn't tell us a heckuva lot about how it's switching, does it? If it's a uniform gap on each loop when you switch it might just be a limitation of the unit. Seems odd they'd let it out the door with that kind of gap though.

[free electron]

Here's the switching part:

MN6631 5 channel CMOS electronic switch
http://www.datasheetarchive.com/pdf-datasheets/Datasheets-110/DSAP009019.pdf

[Rane]

Wow.  My question then is this, would replacing that part fix my problem?  From what I can tell, finding that part could be quite a hassle, and possibly expensive. 

[free electron]

Honestly, i doubt it. Use voltmeter or scope to check the voltages on CTL pins while the unit is switching. Maybe there's one or more switch used for signal muting.
I'd first at least partially reverse eng. the switching part and the signal flow. Then run some 1KHz sin thru it, scope each point and locate where the signal muting happens.

[Rane]

That's what I was afraid of.  I think I'm going to take the easy way out and just sell it on Ebay.  Maybe someone else will have time to fix it, or won't be so bothered by the problem. 

Thanks so much for your help.

:]