Korg DW8000

We got this one in recently, with a dead battery and complete loss of Presets which is not so unusual for a 30 year old synth, and the battery was certainly dead, but the strange thing about the DW8000 is that if the memory is scrambled the synth behaves as if the main CPU board is past saving, if you try to edit a parameter for test purposes and the original value was corrupted, then the synth will crash and reset to program 11, both the parameter value slider or the up and down buttons will do this.

After a new battery was installed in a holder, we tried to reinstate the original programs back into it by means of a Sysex transfer but this just didn’t work, we could see numbers incrementing on the display indicating that presets were being sent over but the results were rubbish. A bit of research showed that this particular 8K had the oldest of Firmware in it, and the possibility that that version didn’t handle Sysex properly (Version 07). We downloaded Wav files of the same factory presets and had a go with that, and even this didn’t work, dozens of attempts being made with differing levels from the computer but with no success. In the end, we dug out an old Cassette player and used the original factory tape that the owner got with the synth and this worked fine, 8 banks were counted on the Green display and the Red displays showed Good at the end.

Of course this synth had other issues apart from the usual battery problems, being a well gigged instrument this one had a beautiful thick line of rust under the keyboard where liquids had been spilled on the keyboard, and quietly ate away at the chassis when no one was looking. A nice drop of something fell through the Joystick and landed on Voice 1, eating through a resistor on the voice PCB. And of course most of the Tact switches are shot, in these mid eighties Korgs they take a lot of abuse along with use, as every function of the synth is handled by the minimum number of switches, 17 in a Poly 61 and 22 in a DW8000 (they have the same switch).


The interesting thing is that a Korg synth of that time (A poly 800, DW6000 or DW8000) behaves as if their CPU boards are completely busted when the onboard battery backed RAM has been corrupted, it can take quite a bit of work to enable the RAM to take a clean new factory bank (or any other), the P800 is not too difficult, the DW6000 is a little more difficult, and the DW8000 is a bit of a pig. There is a routine on power up to initialise the RAM on a DW8000, hold down the 5 and 8 buttons (if they work) on switch on and this will reset the corrupted RAM chip. The synth still produces no sound but at least won’t crash endlessly if the Parameter value controls are used, and you can quickly set up a sound by raising the Oscillator levels (parameters 13 and 23) and then the filter frequency, and the two Envelope generator Sustains. This will give you a sound which you can write to memory to ensure that your new battery is working, and more importantly that there is nothing wrong with the RAM or its circuitry.

The irony of the fact that it took the original factory tape in an actual cassette player (albeit a good one) is not lost on us, the tremendous amount of hassle involved in trying to get the right levels from an allegedly working Wav file to work with this synth from a computer way exceeded the difficulty in replacing the battery in the first place. Our ongoing problems with Prophet 5 tape tranfers led us to recording a Wav file (which is widely said to work) onto cassette tape and to try this out in the future, the P5 is said (by Sequential themselves) to be sensitive to the speed of the tape recorder, and ours has pitch control, another parameter to enjoy !!!.

Actually this is what we did for the P5 , we played a carefully recorded Wav file from a cassette deck into the P5 and it worked first time, giving us 40 working and correct sounding presets (according to the end of the Owners Manual). It also confirmed after a couple of days of testing that our Veroboard memory replacement seems to work very well, and this synth is good to go!

In the grand scheme of things, most Wav files available for most synths on the Net do work, and the all time most difficult was the Prophet 5, and then very recently the DW8000, though we did find the file online to upgrade the DW8000 to the last version, V12 which may have cured the Sysex problems of Version 7, and we did burn and install this, but we didn’t test the scenario, this synth was done, and the next one was on the bench. We don’t have the luxury of exploring every possibilty for every synth that comes in, that is the job of the well equipped hobbyists that populate Forums and proudly proclaim their successes (and the absolute best of luck to those individuals, they are sometimes a useful source of very interesting knowledge), but they are not under such time and space constraints as we endure.

This DW8000 is good to go, the battery change was easy, the rest was not. Apart from the difficulties reloading the factory presets, we had to replace several switches, which meant that the keyboard had to be removed, as did the output Jack panel board because the two front panel boards are connected by a loom that runs behind a plastic pillar which requires removing the output board to get the panel boards out. We also used rust stopping chemicals on the base plate, the kind of stuff you use on cars, scraped the excess rust off, cleaned the area thoroughly and then applied a second coat of rust stopper. This should stop it getting any worse over time. We also replaced the 2 pin IEC connector with the more common 3 pin version, and obviously added a safety Ground, this required filing out the metal power connector bracket by about 1 mm on each side, which if you have the files is straightforward enough, but a little time consuming.

The Polysix yet again!

We get quite a lot of these, and their state of repair varies widely, the range of problems normally encompass the battery, the keyboard and the Tact switches, you will usually get at least two out of three of these. On this particular one, the battery problem had been sorted out many years ago, and while a coin cell wrapped in gaffa tape with wires to the board wouldn’t be our way of doing things, at least it works and there is no damage to the CPU board, R91 was also correctly replaced by a diode so this particular synth retains it’s presets.

This particular Polysix played six voices, and in tune, so it would seem that we had little to do, but about 20 keys of the keyboard didn’t work, and unfortunately there are three different possibiitiies as to why the keyboard won’t play correctly. dirty contacts is the logical first call, and the whole keyboard and contacts were cleaned thoroughly with IsoPropanol and while this made an improvement the results were not satisfactory.. while the keys made contact eventually, there was a noticeable delay between key strike and voices sounding on some keys, and this seemed to vary by how recently the contacts had been cleaned again, as in cleaning them right now gave you a good result, but an hour later, they were no better than they were originally. The top 7 keys of the keyboard could not be made to work at all, despite scrupulous cleaning, and the fitting of rectangular Polystyrene shims under each of them (the keys themselves have a hollow rectangular section which presses down on a circular rubber button, (who on earth thought that was a good engineering idea?) the shims at least provide a solid flat surface to press on the circular rubber buttons and in theory should provide a much better contact with more even pressure, and while this works on difficult keys a lot of the time, it is not infallible. We fitted a contact strip from a Poly 800 onto the top 7 keys, and this made them work perfectly, we don’t know if it just the newer contacts or a combination of the cleaning and the shims and the new rubber buttons, but it did take the new buttons to actually make the keys work reliably. The Poly 800 buttons seem to be made of thicker rubber than the originals and seem to make the keys a little firmer to press.


The above picture is of the shims installed in the keyboard below C3, which was horribly intermittent, and while it worked for a while, even a day later the performance deteriorated to the unuseable. The synth came in to have a CHD MIDI kit fitted so theoretically this was a moot point, but these days a lot of small studios will try to leverage the most out of one keyboard controller, and while the CHD doesn’t do MIDI out, the KiwiSix certainly does, and this does require a fully functioning keyboard.

As of tomorrow, we will have a shipment of Bob Wiegel’s famous contact pads for old Panasonic keyboards, the website address is www.sounddoctorin.com, we will soon see if these will be enough to resurrect a completely knackered keyboard assembly, one of our clients today told us that it did for his Monopoly, with the only problem he had being the extra cleaning of the gold plated contacts on the keyboard PCB for a few notes.

PolySix keyboards do not have Gold Plated contacts.

There I have said it, one of the Holy Grail’s of Internet synth folklore that the Polysix has gold plated contacts on the PCB, which in fairness we assumed to be correct because everyone said so and no one argued or tested this out. Having had so many difficulties with this one, we tried to make the Gold shiny, we have successfully shined Gold plated connectors on Minimoogs, Prophet 5’s and other synths that had them. Using good quality metal polish on the PCB the colour of the keyboard pads went from a dull Gold colour to a shiny Silver type colour, we saw this before on our Fluke Scope when we also assumed that the contacts were gold plated, but turned out not to be, with a major improvement in functionality thereafter. The Fluke also has rubber membrane type switches, so perhaps the Carbon button contacts are a cause, but it could also be from environmental issues and a synth that is more than 30 years old.

That being said, restoring the keyboard PCB to it’s original condition did not make the keyboard work much better, so it’s a kind of race between the sounddoctorin pads or a complete set of Poly 800 type contact strips.

And the winner is Bob Wiegel’s Magic Buttons, we didn’t muck about with just using them on dodgy keys, we replaced all 61 and got an instant good result. All keys working perfectly and with good response times and behaving as well as a 30 year old Panasonic keyboard can. If you have an old knackered out Panasonic keyboard (and you probably do, a huge number of synths used this mechanism in both non-velocity and velocity sensitive versions) then we can definitely recommend these, we will buy a lot more immediately.


The above picture is of the new contact buttons on their carrier, and the six contact block from the synth already has had the new contacts applied, (hence the tweezers).

My Goodness, More Guiness

Actually on this synth (the Polysix) the keyboard was one side of the problem coin, the Tact switches were the other. Surprisingly enough, most of them worked, but a lot of them felt extra sticky with a very distinct click when pressed, which they shouldn’t. Removing the black screws from the front panel and releasing the programmer assembly revealed the horror of the innards, sticky black goo over all the metalwork, the usual dust on the PCB’s turned into a dark brown sludge and some kind of new lifeforms growing over the Arpeggiator Speed Pot and slide switch bodies, a lovely green patina over the legs of a lot of the switches, a picture says a thousand words, so…


The brown staining doesn’t look too bad from the picture, but cleaning it off with cotton buds showed how much there was and how dirty the pcb got, we did the Kiwisix LED mods at this time because we had the switches out and taking out a few more components made cleaning the pcb a bit easier and more successful. The switch bodies themselves were brown and sticky, and to get to the point we replaced all 19, two new keytop carriers being necessary as they were broken, and one keytop (which was missing) which came from a TR808, but which did fit quite well in the front panel, and doesn’t look too much out of place.


We tend to fix first and Photograph later, so it is not too common for us to have before pictures, but here is another example of the staining on the metalwork and pcb’s, it actually doesn’t look too bad from the Photos, but the reality is much worse than the pictures display, but in the end the complete replacement of all of the switches and the thorough cleaning of every part did fix this issue.