So the PCBs for my first build have arrived (yay!) and two of the four boards worked without bodges (sorta yay!). The backplane seems okay and the VIA board worked without issue. The main board (65C02, EEPROM, RAM, some decoding) had a problem I knew about before the boards arrived and which was fixed with a couple of tiny bodge wires.

But the serial board? Not so good. There's not much on it. A couple of chips for decoding (working), the 6551 with its crystal and cap and that's largely it. It's the same config I had on the breadboard, which was working fine there.

The serial port sends out data fine, so the 6551 is working to some extent. But there's a problem receiving.

I checked the RX line and it's being held at 5V normally with the signals dipping to 0V as expected, so no obvious problem with incoming signals. However, the /IRQ signal from the 6551 is being weird.

I have a 10K pullup for the /IRQ line on the main board. Sure enough, it's normally at 5V - albeit with some 50Hz ripple. But when data comes in on the RX line, the signal drops to around 3.3V – and stays there until I reset. It never drops even close to 0V and therefore the serial routines in the ROM are never triggered.

I tried a different backplane, with closer spacing of the boards - exactly the same deal.

Any idea what I can probe or bodge next? Is the 10K pullup too strong?

_________________
I like it when things smoke.
Blog – Zolatron 64 project

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

Oops, while I was typing Garth covered most of this. But...

Either the /IRQ line on the main board is harder to pull down than it ought to be, or the 6551 and its connections are weaker than they ought to be in regard to pulling down. Here are some of the possible explanations.

/IRQ line on the main board is harder to pull down than it ought to be:
- shorted to another signal line, perhaps -- one that's trying to be high?
- what happens when you use another 10K resistor to pull it low? Is the result Vss/2 as expected?

6551 and its connections are weaker than they ought to be in regard to pulling down:
- bad solder joint in the IRQ circuit
- the Gnd pin on the 6551 itself is not at ground. Measure it, right at the pin itself.

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

I am indeed using the 'S' version of the W65C22 on the VIA board.

I haven't had much time to run tests, but I thought I'd briefly try unplugging the VIA board to see if the /IRQ output on that was holding the line too high. Weirdly, the serial board stopped working completely! No output. Gah!

The /IRQ pin on the 6551 is still at 5V, but still with lots of noise. The GND pin is at 0V.

I will try other things tomorrow.

_________________
I like it when things smoke.
Blog – Zolatron 64 project

Thanks for all the tips. In the end, Garth, it was your primer that tipped me off to the solution. I just don’t think the 6551 could fight against the totem pole output (a phrase I had to look up) of the 65C22S. I switched to an R6522 and it all worked. Well, there was a lot of other cussing, bodging and probing along the way, but if you want a fuller story it’s here: https://mansfield-devine.com/speculatri ... -it-werkz/

It’s still a little weird in that I swear these are the same chips I had in my (working) breadboard version, which had an even stronger (1K) pullup on the /IRQ line. Oh well…

_________________
I like it when things smoke.
Blog – Zolatron 64 project

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

Thanks Garth. There will always be another spin of the board coming along, so I’ll build that in.

_________________
I like it when things smoke.
Blog – Zolatron 64 project