Problem possibly due to midiSpeak

[napierzaza]

Is it normal to have the keypad a little messed up after installing Midispeak? I installed it in a push-button Speak and Spell and now I have some buttons that trigger incorrectly.
Column one is the button pressed and column two is the trigger.
E - C
J - H
O - N
T - R
Y - W
Enter doesn't work (presumably it's triggering module which does nothing.

If you know about the push-button variety you know there are two matrices inside, and EOY and JTEnter are the far right hand sides of both of those matrices. Can you tell me if this is a normal effect of MidiSpeak or something else?

I though it might have had to do with me stuffing in the module, but that would only make sense on the left hand side where it's located.

Thanks.

[John]

A proper MIDISpeak installation will not have any impact on the function of the host Speak. Your trouble is almost certainly caused by a wiring problem. Keypad "aliasing" is likely caused by a short circuit or mixed-up connections.

I would double-check the wiring against the connection diagram, and also use a continuity tester to check for shorts between the wires. You might also disconnect everything, then connect each wire one-at-a-time to see which one is causing the trouble.

You're also welcome to post a photo for inspection, but that doesn't always work too well. ;)

Hope this helps--please let me know your results. :)

[napierzaza]

Hi John,

I'm afraid you're somewhat wrong.

I installed the cables as follows (original push button with the same model of circuit board).

http://highlyliquid.com/kits/midispeak/pcb-pushbutton-1.jpg

When I disconnect the keypad 1 and keypad 3 the buttons resume their proper assignment. I'm sure the pads I connected to are correct, so I'm not sure why this is happening.

Thanks

[John]

Hi John,

I'm afraid you're somewhat wrong.

I installed the cables as follows (original push button with the same model of circuit board).

http://highlyliquid.com/kits/midispeak/pcb-pushbutton-1.jpg

When I disconnect the keypad 1 and keypad 3 the buttons resume their proper assignment. I'm sure the pads I connected to are correct, so I'm not sure why this is happening.

Thanks

Ok. Those particular connections (Keypad 1 & 3) are only there so that the MIDISpeak can reset the host's "sleep" timer (to prevent the host from turning itself off after keypad inactivity). So, without them, you should have normal MIDI functionality.

Is the MIDISpeak working correctly? (...do you get repeatable sounds in response to MIDI notes?)

Does the display appear normal, aside from a brief "flash" when a MIDI note is sent to the unit?

If everything else is working correctly, then we'll continue to pursue the keypad issue. If not, then there may be other issues with the install.

[napierzaza]

> Ok. Those particular connections (Keypad 1 & 3) are only there so that the MIDISpeak can reset the host's "sleep" timer (to prevent the host from turning
> itself off after keypad inactivity). So, without them, you should have normal MIDI functionality.

Hrmm, but it does seem like a useful feature.

> Is the MIDISpeak working correctly? (...do you get repeatable sounds in response to MIDI notes?)

Yes, I was able to trigger notes over midi, it was the same each time, but I didn't notice the sound changing with each different key. But I have to read more about how to control the actual unit.

> Does the display appear normal, aside from a brief "flash" when a MIDI note is sent to the unit?

Yes, it is working fine, going into "underscore" mode fine as well.

> If everything else is working correctly, then we'll continue to pursue the keypad issue. If not, then there may be other issues with the install.

Okay, well it appears as if everything is a-okay on my side. I've checked my interconnections and soldering a few times over so I'm pretty confident.

[John]

> Ok. Those particular connections (Keypad 1 & 3) are only there so that the MIDISpeak can reset the host's "sleep" timer (to prevent the host from turning
> itself off after keypad inactivity). So, without them, you should have normal MIDI functionality.

Hrmm, but it does seem like a useful feature.

> Is the MIDISpeak working correctly? (...do you get repeatable sounds in response to MIDI notes?)

Yes, I was able to trigger notes over midi, it was the same each time, but I didn't notice the sound changing with each different key. But I have to read more about how to control the actual unit.

> Does the display appear normal, aside from a brief "flash" when a MIDI note is sent to the unit?

Yes, it is working fine, going into "underscore" mode fine as well.

> If everything else is working correctly, then we'll continue to pursue the keypad issue. If not, then there may be other issues with the install.

Okay, well it appears as if everything is a-okay on my side. I've checked my interconnections and soldering a few times over so I'm pretty confident.

Can you post a high-res photo of the install?

[napierzaza]

http://inventory.concordia.ca/DSC03455.jpg
http://inventory.concordia.ca/DSC03456.jpg
http://inventory.concordia.ca/DSC03457.jpg

Here you are. Can't keep them up forever.

(note from moderator: adding images as attachments)

[John]

Here you are. Can't keep them up forever.

Thanks for posting such excellent pictures. :) I have them stored, if you need to delete them.

I agree--I can't seem to find anything that looks out of place. My next step would be to remove the batteries (you are using battery power, correct?) and use a continuity tester to check for a short circuit between either of the keypad connections and any of the other pins on the keypad matrix.

Is it just one connection or the other (of keypad 1/3) that causes the key aliasing to occur?

I will also see if I have a pushbutton speak on hand and examine it to see if I can gain some insight to the problem.

[napierzaza]

Hi, I already continuity tested the button pad, I will check the rest. I didn't see if one or the other corrected it, just both. I'll let you know (in like 5 minutes).

[napierzaza]

Okay,

It's the YELLOW keypad wire. I removed the yellow and it worked, re-attached it and removed the blue and it didn't work. So it looks like that's the wire. Also there are no shorts between the keypads.

Oh, and why can't you use and adapter? I realize that it's not wired for that, but what would be needed? I kind of plan on using a PSU eventually.

[napierzaza]

Bump.

Any news?

[John]

I've located a pushbutton Speak and will post an update within the next day or so. Thank you for your patience.

Regarding the power supply/battery question, I've made a new thread here:

http://forum.highlyliquid.com/showthread.php?t=91

[John]

Hi again, thanks for your patience.

The Speak I found was not in working order, so I'm not able to attempt to reproduce the problem. I'm out of ideas at this point--sorry for not having an answer for you. You're welcome to return the MIDISpeak if it does not meet your expectations.

[napierzaza]

Hi John,

Thanks for the answer, it shouldn't be a problem, I will keep it.

Those keypad contacts do what exactly? Are the triggering keypad presses or something?

[John]

Hi John,

Thanks for the answer, it shouldn't be a problem, I will keep it.

Those keypad contacts do what exactly? Are the triggering keypad presses or something?

Yes, those two wires simulate a keypress of the "clear" button. This is the way that the MIDISpeak prevents the host from automatically switching off after several minutes of keypad inactivity.

[napierzaza]

Hey John,

Can you explain to me how I might trace that button? I was trying to test the wires from the button pad but I don't get any kind of continuity between the wires when buttons are depressed.

[John]

Hey John,

Can you explain to me how I might trace that button? I was trying to test the wires from the button pad but I don't get any kind of continuity between the wires when buttons are depressed.

I think your approach is correct. There is still some resistance even when the key is pressed--perhaps it is too large to trigger your continuity tester. You might just monitor a resistance reading instead...