Nord User Forum

[Marlowes]

Not too many helpful hints I'm afraid. The schematics are nowhere to be found, so luckily I was able to borrow a second, mostly working, unit so I could probe around with an oscilloscope and learn how the working one worked and pinpoint where the broken one wasn't.

The other unit was a later revision, and interestingly Clavia had changed the power supply connector from a standard 3.5 mm connector to a telephone type RJ-11 6/4 connector. Inside the changes are minimal, but I noted that in Michael's unit the markings on most of the chips had been unhelpfully scraped off by the manufacturer, but I suppose they realized that that didn't accomplish that much, so on the second unit they were all intact, which made it easier to figure out what was going on.

I don't know if the fault on Michael's unit is typical, anyhow, an output on a 4069UB hex inverter had ceased to function as an output (it behaved as though it wasn't connected to anything on the chip at all). Either it was due to old age (not unheard of with this generation of CMOS chips), or something in this region in the circuit taxes that chip too hard causing premature failure - this part of the chip is used as a voltage controlled oscillator for the sample clock, which is not really the intended use for a digital chip, although it is common practice.

An interesting note I made is that contrary to most 8-bit drum machines of the period (such as the Sequential DrumTraks or the Oberheim DMX and DX) which use µ-law compressed audio which compresses 12 bits of audio data into 8 bits for storage in an 8 bit EPROM chip, the DPP-1 uses an 8-bit linear digital-to-analog converter and hence the samples are stored as 8 bit linear data.

The actual cartridge simply contains a single 32 kbyte EPROM chip, and the two switches on the front of the machine select one of four regions in the chip, in effect working as a sound selector. It would not be hard to program your own sounds if you have an EPROM programmer, but unfortunately it looks like there's not enough space for an IC socket inside the cartridge so it would not be as straightforward as removing the chip, programming it in an EPROM programmer, and replacing it.

Anyway, I suspect a lot of the charm of a machine of this vintage comes from the original sounds. And I think the 8-bit microcomputer (the ABC80) used to create them is still on exhibit at the Clavia factory in Stockholm, together with the specially designed hardware for the sampling process.