MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Esq-1 keyboard repair
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Esq-1 keyboard repair
pre55ure
Looking for advice and opinions on this one.

I have an ESQ-1. 2 sets of 8 keys aren't working.
I found a schematic of the key matrix online and was able to get a better idea of what was going on. It turns out that there are 2 "rows" that aren't working right. One of them just doesn't have any continuity between the depressed key and the header pin, and the other one is very strange- if I use my continuity tester, it will beep, but it takes a few seconds, almost like its charging up seriously, i just don't get it . When I just look at it with the normal setting on the MM it starts with a high resistance and then drops down over a few seconds till it settles in at about 10r. This seems totally bizarre to me, could this be related to old key contacts and oxidation?

So the real question... Is this even worth bothering with at this point or should I just turn it into a rackmount module?
The keyboard seems like a nightmare to take apart, and even after I get it apart I still don't know what is up with the weird variable resistance contact.
Thoughts?
Suggestions?
fuzzbass
I'm not certain about ESQ1, but some Ensoniq keyboards consist of two PCBs with sensors, joined with ribbon cable. This connection is root cause of many keyboard issues due to failing mechanical contacts in the ribbon cable or headers. If ESQ1 is like this, and this has not yet been replaced, it should be.

I made this change to a VFX. You do have to disassemble the entire keyboard to access the area, but it is not a nightmare, and it goes back together just fine.

The VFX keyboard did not work by electrical contact, but rather proximity sensing. There were little metal plates on the undersides of the keys, and sensors on the PCB.
jhulk
esq-1 uses fatar keybeds they are just a spring with 2 bars note on note off

they use a diode matrix and have a special 8bit cpu keyboard scanner chip

that does a velocity from length of pulse of the diode being that when hitting a key harder the key is held longer

this is all done in software which is coded into the masked rom of the 8bit cpu

vfx is totally different as this uses a magnetic coil boards and there are 2 of them but it uses these magnetic feilds as a control for poly aftertouch
in early boards they used to use cheap ribbon connectors which would oxidize and cause them to fail

when doing warranty service on these the bulliten was to hard wire them which solved the problem on these they use special 16bit keyscanner chips the magnetic coils have a jfet for each key usually when doint the calibration the jfets are switched by the scanner cpu to test if its receiving the magnetic signal if a jfet has failed it then cant calibrate

on the matrix of the esq1 it has a send for each key range all c, c#,d/ and so on for the 12 notes and then a return from the end of each octave which is a resistor divider network so if c2 is pressed the it will go through the divider network to its diode which will return via thats return for that octave so i would say you have bad returns for the octaves missing you need to check the ribbon cable from the motherboard to the keyboard pcbs to see if you are gettin continuity if you are you could have a faulty keyboard scanning chip

i have spares of both if you need then they are also the same keyboard used in the mirage and emax1/2
Dave Kendall
Hi.

I would certainly suspect header pins and connectors in older Ensoniq products. Several problems in my old EPS (bought new) were solved by replacing and/or cleaning them.
One bad batch was actually corroded and green when the connector was removed for inspection. . .

Ouch.

cheers,
Dave
pre55ure
jhulk wrote:


On the matrix of the esq1 it has a send for each key range all c, c#,d/ and so on for the 12 notes and then a return from the end of each octave which is a resistor divider network so if c2 is pressed the it will go through the divider network to its diode which will return via thats return for that octave so i would say you have bad returns for the octaves missing you need to check the ribbon cable from the motherboard to the keyboard pcbs to see if you are gettin continuity if you are you could have a faulty keyboard scanning chip

i have spares of both if you need then they are also the same keyboard used in the mirage and emax1/2


Cool, thanks for the replies everyone.

Jhulk- I might take you up on one of those spares.

I tested the cable and it's good. So it's either the chip or the header pins (though the weird "charging up" behavior makes me think it's likely the chip, cause I cant see how the pins would cause that behavior.)

Unfortunately it appears that the chip is on the other side of the pcb. So it looks like 61 keys need to be dissasembled before I can get to it. Dead Banana

Any tips on how to clean off the metal parts that make contact? If I have the whole thing dissasembled I might as well get some of the grunge out.
pre55ure
Ok, so opened it up and theres no chip.

Just a bunch of dirty looking pcb traces.

So... any ideas on the next step? i'm thinking of just trying to reflow all the solder joints, cause I can't really think of anything else to do.
jhulk
the chip is on the mainboard the cable that comes from the main board to the keyboard its socketed idc connection the chip is socketed its there large ic i think its a 40pin but will check on monday in my stock inventory to what size it is

i would check continuity from the pads of the idc connection on the mainboard and the keyboard while the idc ribbon is attached then from the 40pin ic and the keyboard idc connection to see if the chip is getting the signals i/o

the pcb under the keys just has resistors and signal diodes

what i would do is press each key where you are losing contact to see if you are getting a voltage from the diodes

which will let you see if they are triggering
jhulk
ok i have a spare board in front of me

its u1 its a its a rockwell 8bitcpu mask rom r1120-12 and its a 40 pin chip its right next to the battery and the keyboard connector idc 26 is j2 this is left of the chip and interfaces with the chip
Dave Kendall
From memory, the service tech in Putney, London (back in the day when ensoniq actually had local service centres!) used isopropyl and very mild abrasive for mild cases of gunk on pins, and simply swapped out the horrible green-corroded multipin connector. - including housings, in which the contacts weren't accessible.

Mild abrasive seems like rough treatment to me, but that probably depends if the pins are plated or not.
.
IIRC, the EPS mk 1 was very similar in build to the SQ80 that superseded the ESQ1, and was not very reliable. Don't have it any more, so sorry I can't check details.

Good luck smile

cheers,
Dave
BruttoBello
Pre55ure, you had the ESQ-1 open right? How hard do you think it would be to turn it into a rack mount/ desktop module?
Thanks,
Nick
jhulk
VERY EASY ITS ONLY A FEW PCBS
estin
bumping this old thread as I have an ESQ-1 with a row of 8 dead keys. Ever find a solution? I have tried everything short of replacing the keyboard controller IC (since they no longer exist)

as a side note for irony, I decided to hook up my keystep to the esq-1 so I could play it, and the keystep has a dead key very frustrating
wally01
[quote="estin"]bumping this old thread as I have an ESQ-1 with a row of 8 dead keys. Ever find a solution? I have tried everything short of replacing the keyboard controller IC (since they no longer exist)

Have you tried re-flowing the solder joints under the keyboard controller IC?

My ESQ-1 had intermittent startup issues. Re-flowing the pins under the 6809 processor fixed it. After all these years the solder joints are starting to fail with the thermal expansion and contraction.
Finding a replacement keyboard IC is next to impossible. I'm looking for the keyboard scanner IC for an SQ1-Plus.
estin
[quote="wally01"]
estin wrote:


Have you tried re-flowing the solder joints under the keyboard controller IC?




This I have not tried and would be a good idea. I did do this on the keybed PCB ribbon connector end with no improvement. I didn't think to pull and inspect the main board area.

I feel like its unlikely to be the IC itself as the rest of the keybed works...


when inspecting the keybed PCB i noticed a couple corroded traces so i jumpered them. Not only did it not help but it created a new problem where one of the previously dead keys would trigger the sound of a number of keys all at once only when the key was released
it still made no sound when pressed down. I double checked all the connections and nothing was bridged or incorrectly jumped to the wrong spot. I removed the jumpers and it went back to just the 8 dead key issue.






as a side note i did succeed in fixing the arturia keystep so I can at least play the ESQ-1 right now. It's peanut butter jelly time!
wally01
Good to hear you got the keystep fixed.

If you lost 8 keys i'm guessing the problem is on the keybed pcb.

The keybed has bus bar conductors that run the full length of the keyboard.
The bus bar is connected to the keys in groups of 5 , 4, or 8 keys.
I'm guessing there is a bad connection on one of the 8 key groups.
What section of the keyboard is affected?
estin
it effects G# through D# just to the right of center keyboard. I am familiar with the bus bars you mentioned, I re-flowed all of their connections to the PCB. The only note worthy thing I noticed was one set of the bus bars middle connection was bridged (intentionally) to join the two sets. None of the other sets showed this. I believe this was near the troubled keys as well. None of the other bus bars were bridged this way. Should I remove the bridge?
estin
I just checked and the bridged bus bar section is correct, there are two done this way as well.
wally01
That section of the keyboard has 2 groups of 4 keys connected together.
So yes - the 2 center pins are jumpered and the 2 outer pins are jumpered.
Glad i took a photo after i repaired my keybed.
estin
wally01, thanks for encouraging me to take another look at the keybed PCB. I now have a fully restored and fully functional ESQ-1!!! Rockin' Banana!

after probing with the continuity checker I found another bad trace on the key bed PCB. I also ended up re-flowing both ribbon cable connections, and the keyboard IC holder on the main board just for good measure.

One more 80's relic alive and dropping some dope lo-fi pads MiniMoog


here's a few pics of the jumper work needed.



wally01
Congrats!
Glad to see it's repaired. It's peanut butter jelly time!

Cheers
Pat
It's peanut butter jelly time!
estin
Thanks Pat! Guinness ftw! restoring this synth has been a marathon but what rush now that its done. I appreciate all the help and guidance you provided.
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group