Bidirectional positive-logic to negative-logic converter?

[jayeye]

I am trying to interface TTL circuitry to an old piece of equipment (a Texas Instruments PC-100C printer cradle). That one has two supply rails, at (approximately) -10V and -15V; logic 0 is, of course, 0V, logic 1 is -10V on some and -15V on some others. If these were unidirectional lines, I could do the trick with just some variant of max232, but they are not, they are bidirectional. If they had only one supply, I could invert the TTL logic, make a level converter to, say, 10V, then connect the TTL Vcc to the old equipment's Vss and power the TTL circuit from a battery. Digging around, I found a patent to convert TTL to ECL levels, but (a) that appears too complicated (ECL working in the active region of transistors and all that), and (b) not sure it will actually work for bidirectional signals.

Help?

[T3sl4co1l]

What kind of bidirectional signals? Is there a direction flag (typical of controlled buses) or is it open-collector or what?

How much current do the signals provide?  Are we talking old fashioned PMOS (few mA)?

Do you have access to Vss?

Tim

[jayeye]

Yes, I have access to Vss. The schematics call the -10 and -15 (or thereabouts, I don't have them right in front of me) Vgg and Vdd. The two-phase clock is one of them, the rest of the signals are the other. Not sure if they are open-collector (well, open-drain, while the circuitry is almost certainly cmos, it needed to be compatible with the very old SR-52 calculator which was PMOS, hence the weird voltages). There is no explicit direction indicator signal, but I'm sure I can reconstruct one by counting clock cycles from another signal (that's how the equipment does it).