Re: Rowe R87 works perfectly, then forgets where home is.
Posted: Thu Mar 10, 2016 4:27 pm
Rob, sorry about the name mix-up, I thought this was a follow up question from Ron.
Around the time of this R87, I used to design microprocessor based computer boards, so I am familiar with designs of this era, but I wouldn't have done it the way Rowe did. It looks to me like a bunch of analog designers were just getting into digital design and so their techniques are a bit strange. They use transistors, diodes and quad comparator ICs to strobe various inputs and put them onto a common data bus. There were plenty of bus driver ICs available that did the same thing, and would take fewer discrete components and make more sense to people troubleshooting today. It took me quite a while to figure out what is going on, especially without a schematic. I had to put the PC board on a copy machine and then follow the traces with a pen to reconstruct parts of the schematic. What a pain!
The one thing that was clear to me early on, is that the problem had to be somewhere between the Home LED on the Mechanical board and the diode next to Q116 on the CCC. The other side of that diode is a shared data line. So if other things tied to that same data line, like the Most/Least Favorite switch, worked, then the microprocessor was fine. I also doubted it was a memory problem because I could read out Most and Least favorite on the display. Yes, only a segment of memory could be bad, but that was a long shot.
One thing that bothered me about this design, is that the Home and Index signals both drive a transistor on the CCC. The resistor on the base of those transistors is on the Mechanical control board, while the transistor is on the CCC board. That basically exposes the base of the transistor to any noise on the cable harness. I would not have done it that way.
I have one woofer out and yet this machine sounds as good or better than any I have worked on. I really like the light-weight tone arm. It looks what you'd fine on a high end turntable, especially when you compare it to, say, a Rock-Ola of the same years.
Around the time of this R87, I used to design microprocessor based computer boards, so I am familiar with designs of this era, but I wouldn't have done it the way Rowe did. It looks to me like a bunch of analog designers were just getting into digital design and so their techniques are a bit strange. They use transistors, diodes and quad comparator ICs to strobe various inputs and put them onto a common data bus. There were plenty of bus driver ICs available that did the same thing, and would take fewer discrete components and make more sense to people troubleshooting today. It took me quite a while to figure out what is going on, especially without a schematic. I had to put the PC board on a copy machine and then follow the traces with a pen to reconstruct parts of the schematic. What a pain!
The one thing that was clear to me early on, is that the problem had to be somewhere between the Home LED on the Mechanical board and the diode next to Q116 on the CCC. The other side of that diode is a shared data line. So if other things tied to that same data line, like the Most/Least Favorite switch, worked, then the microprocessor was fine. I also doubted it was a memory problem because I could read out Most and Least favorite on the display. Yes, only a segment of memory could be bad, but that was a long shot.
One thing that bothered me about this design, is that the Home and Index signals both drive a transistor on the CCC. The resistor on the base of those transistors is on the Mechanical control board, while the transistor is on the CCC board. That basically exposes the base of the transistor to any noise on the cable harness. I would not have done it that way.
I have one woofer out and yet this machine sounds as good or better than any I have worked on. I really like the light-weight tone arm. It looks what you'd fine on a high end turntable, especially when you compare it to, say, a Rock-Ola of the same years.