Hammond SK2 keyboard horror show

[Randall]

A friend brings me a nice Hammond dual manual two tier keyboard with 16 keys dead on lower left and 8 keys dead on upper right. Hammond was kind enough to send me the service manual, and with some redundancy I was able to narrow it down to the keyboards themselves. Problem is, Hammond/Suzuki doesn't make them, an Italian company Fatar does. The keyboards are connected to two boards under the keys that have the contact switches in them, one for the left half and one for the right half. There is one part number for the left side and one for the right side. I swapped boards from upper and lower right, and the dead keys followed the board, so I will assume the same is true for the left sides.

Hammond seems to consider the entire keyboard assy a single part number, which sucks. I get nothing in a Google search, and no indication they are replacable from the manual. So far, no answer from them if those boards are obtainable. They use them on several models. SO, now the guy may be looking at a very expensive repair on a $2.5K keyboard. But as if that wasn't enough...

The support guy who sent me the manual recommended I make sure it had the latest updates, which is on their website. You have to wipe a USB Flash drive, stick it in the Hammond where it gets formatted with a simple file system. Then after downloading the update, unzip it and copy all the system files into the system folder on the USB drive. Once that is done, you insert it into the powered off Hammond, and power back up while holding 3 buttons. This is supposed to initiate the update. But, do you think I could get it to work? Over and over and over I tried. I tried different Flash drives, I read everything I could find, I went to different Hammond support sites on different continents. All day off and on I tried, and all I got was "no update file" when I initialized it.

Eventually, I came across a support page in Europe that had a different, more recent update file on their download page, it took me a while to figure out how to actually download it, but when I did, it worked! So Hammond USA, as well as Australia and a couple of others I tried had the same old and not working update file on their sites! That's all it was all this time. Goddamn it.

I got the keyboard updated finally, and it did not correct the original problem of dead keys, but at least I can speak with more confidence that the issue is indeed those boards, and not a firmware issue, which I know they will want to maybe blame it on.

It's odd that they died in groups of 8, since the key mapping section works that way also. But, I'm pretty certain if dead keys jump from top to bottom with a board swap, then that's the cause.

[The Dude]

I can't see what you're working with, but it's odd and rare that that many keys in a row would be bad. Can you post some pics? I've worked on lots of keyboards (not this one), and never had to order a whole board. Usually it's something got spilled in the keyboard and the switch contacts need a good cleaning. Or, the little rubber contact strips are worn out. Does this have the little rubber contact strips or actual metal key switches? I do remember encountering a similar problem on a keyboard where an entire group of keys didn't work. A liquid spill had eaten through some traces on the key contact circuit board. I was able to jumper the traces back together and salvage the board.

[Enzo]

Actually groups of keys going dead is fairly common. Most common is one or two random keys because the contacts go weird. But the keys are wired in a matrix. Rows and columns. And if something hangs up a row, then a group of keys are stuck. Or a column and something like every 8th key goes dead.

Fatar was and maybe is a huge maker of keyboards. Most if not all old Ensoniq keybeds were Fatar.


Updates unlikely to solve a dead keys problem, as the old software worked before.


If you can verify the main board circuits work by board swapping, and it is on the keybed boards, then what I generally do is just get out my meter and go down the rows of diodes and just check each one real quick. Also, if like most, the connectors on the boards either each have a wiring harness connector to the main board, OR some of them have one connector for back to the mains, and just short jumper cables between the two halves on the bed. In either case, look VERY carefully at the solder on the connector pins. All the constant key pounding sends vibrations through the keybed, joints break. One row or column line loses contact and that whole group go dead.



Some questions. Are these mechanical contacts or the soft rubber kind with the black dot that touches printed contacts on the board? Really old keyboards used things like stiff piano wire contacts, even gold plated ones. Metal blade contacts like Yamaha DX7 and the many Korg models that used that DX7 Yamaha bed. There were the contacts that were silver plated springs stretched across a contact. Got a photo? Also are these simple single contacts - like an organ? Or are these keys velocity sensitive, and so have two contacts per key?

[Randall]

Victory. One of the first things i did was run down the string of diodes, and they all acted the same, beeping at .6 volts. There are also 6-7 resistor looking things that only have a single black stripe.They all measure .03 ohms, I have no idea what they are or what they do. I had a strong hunch that the groups of 8 was a clue, and it turned out to be right. These boards are the rubber over carbon (I think) contacts. The rubber came off fairly easily to reveal the common from each octet of switches was shared as it goes back to the connector. Some of those traces were fouled for some unknown reason, and were no longer conducting. Maybe perspiration, I don't know, but I saw no evidence of spillage. Jumping across the bad spots got the dead keys working again. The board with only 8 dead keys wasn't as bad as the one with 16 dead keys, so for that one I jumpered all 4 of the traces that went thru the bad looking spot. I figure it was just a matter of time before more of them fail, and I really don't want to have to open this up again, my shop is too small to have keyboards lying filleted for days.

I wonder if it is a part defect and not a spill, because even when I scratched away enough to get to clear copper, it was a challenge to solder to it. The copper was quite resistant to this, and I didn't get quite the secure connection I usually would want, but in this case I took what I could get.

I hit the wires with a dab of Gorilla glue to keep things steady, and tomorrow I will button this up with a good feeling. I saved my buddy some dough, knowing there is no one else around here he could have turned to. All systems are go.

Oh, and for the record, I did get an email from Ray at Hammong telling me he did have some of those boards, but I'd have to go thru a dealer to get a price, etc. Good to know there were options if I didn't get these repaired.

An extra glass of wine for me tonight!

[Enzo]

Little "diodes" with one black stripe are zero ohm resistors. They are easier for the pick and place machines to handle than plain wires. So they are just jumpers.

It can be hard to remove the green solder mask from traces. I really doubt the problem was a defect. Spills don't always leave big brown stains.


You would know right away if you got it wrong, but observe the dual contacts. One is deeper than the other. So when the key presses down, one closes before the other. The system detects the difference in time and interprets it as speed.


To clean traces I have great success with a fibre brush eraser.

https://www.eraser.com/products/fibe...lass-eraser-2/

[Randall]

Enzo, I think you misread. The diodes checked out okay. Not sure what other "diodes" you are referring to. 0 ohm resistors are news to me, but it makes sense after you point it out.

I used a fibre eraser like the one you show. It wasn't difficult in this case to clean the traces. I even cleaned it with alcohol. It was difficult to get solder to wet and bond however, even a couple of inches out from the bad spot.

[Enzo]

You called them resistor like things, to me they look like diodes with the single stripe and all, so I called them that. That is why I put it in quotes. . In any case, they are just zero ohm resistors.

[Lars71dk]

The support guy who sent me the manual recommended I make sure it had the latest updates, which is on their website.

Hi Randall, I have issues with my SK2 as well, great you got your fixed, I will give mine a go. Would you mind sharing the service manual?
Thanks!