So the service manual says the collector of Q15 should latch high when "start" is pressed, and low when "stop" is pressed. Similarly, assuming Q15 is already low, it should latch high when "stop" is pressed and "low" when "start" is pressed.
The problem is that the circuit works fine when it's start then stop, but when it's stop first, Q15 doesnt latch. I've checked both 33k resistors to make sure they're in tolerance and they're fine. Anything else I should look at?
Can we assume that all arrows indicate connection to a common logic supply?
You'd need to see the rest of the circuit, if something else was supposed to be capable of pulling the collector of Q15 low, as the circuit, by itself, is not a latch. Latching requires involvement of the outputs of the other circuitry, which isn't included in your image.
The condition of the voltage at Q15's collector cannot affect the functions of the stop and start switches, in controling the function of Q15.
( There are two R78s in the drawing... so it can't have been proofread or checked very carefully.)
The output from the "Schmitt trigger" is used to activate the "CE" line of the cmos SRAM IC in "write" mode. I can't see any other connection from the output to the input that affects its operation.
The purpose of the diode/RC network hanging off pin 6 of the 4011 flip flop is unclear to me.
Good question when it stopped working. I grabbed it off eBay for around $30...vintage ones in good condition often sell for hundreds as techno/electronic musicians often fetishize this old stuff. Check out what the big brother of this box from a few years later, the TR-808, sells for on the vintage market sometime...
This one was in bad shape, looked like it had been sitting in a wet basement for years and had water/rust damage. Only a small corner of the PCB was corroded fortunely, though there were a couple obviously lifted traces which I jumpered.
It also had quite a few failed parts...IC 4011 was shot, the FET in the accent control was dead, the switches were frozen and needed to be freed, and the contacts in the push buttons were worn out and needed to be replaced.
It is essentially functioning normally now in "play" mode, but I can't write patterns due to the RAM not being advanced/written correctly by that circuitry at the bottom. When one tries to use the buttons to advance the steps the box just lets out a blast of noise.
Here is the quote from the service manual about that section:
" 3 SCHIMITT TRIGGER (Q15, 17, 18) This circuit functions in WRITE mode only. The collector of Q15 goes H when either the START switch S5 or the STOP switch S6 is turned ON, and goes to L when the switch set to ON is turned OFF."
Viewing the larger drawing, there are two types of skyhook supply arrow indicators. One is solid arrow off of C13, the other is skinny arrow off of C48.
-The skinny connection is a switched contact (input socket plus volume control S8 on VR2) to a ~6V battery pile, running most control logic.
-The solid connection is a filtered skinny supply running to analog signal sections. While the TC5501 (IC1) runs from skinny-minus-a-diode-drop on C39, the rest of the control circuit runs on the unreduced skinny rail. This means that input signals to IC1 can (potentially) exceed it's VCC pin. This includes some direct switched contacts on S3 and some CMOS drive outputs (~600R) of IC2 and IC3 - not good. The rest are resistive pull-ups of 10K or greater - not an issue. It would not be easy to predict what response to expect from IC1, when inder the infuence of or subsequent to the application of >VCC voltages on the affected pins. Supposedly it has worked as expected since 1980, until it didn't. Any idea when this (didn't) was?
Changing state of the Q15 collector node in the "write" mode is also expected to produce a R/W pulse on IC1 input pin 20. Expect this is negative-going from about 3/4 of logic supply.
There is nothing in this circuit that could hold Q15 collector low, externally in "write" mode; the CE2 line of IC1 and the E line of IC2.
In "write' mode the clock and FF circuitry are effectively out of the action. Q15 can only respond to button presses, directly and in real time. No latching should be expected, as drawn.
OK. A confusing description of the switch functions.
There is some funny drawing of connections around the footswitch socket J3, so that it's not obvious what plug-free operation is expected. Hardwiring the output of IC3 pin 4 to J3 pin6 would temporarily overstress the IC every time the start switch was pressed. If this connection is expected, then a latching function could be expected of the start switch alone.
Q15 is normally saturated. It should turn on shortly after application of power and remain on. (C41 effect)
Without the shorted wiring in J3, both start and stop switches should produce a temporary turn-off of Q15, allowing a longish positive-going pulse on it's collector, and a shorter negative-going pulse on normally-off Q16 collector (IC1-R/W).
If Q15 is off, with collector high for any length of time, then you would need to check what is preventing it from turning on normally.
With the J3 connection, a start button push will latch Q15 low without producing a R/W pulse, and a stop button push will unlatch Q15 to produce a R/W pulse.
If that's what you see on the collectors of these two transistors, then there's nothing wrong with this section, if amplitudes and timing are CMOS compatible. You need to scope with high-z x10 scope probes(~10meg z).
possible faults R85 hi z Q18 short/leaky R87 hi z Q17 leaky C45, C43, C41 leaky or backwards
After high humidity storage, you should check higher-value resistors for drift and all preset resistors for wiper and terminal contact.
You must use real 15V 4011 types for IC3, not HC or other 5V logic types, when replacing it.
I'm not sure what sort of noise is expected when single stepping the device, but with S4 in the off possition, you shouldn't hear anything.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.