so… I am designing a busboard for the sync bus connections on a Cadet system.
this is what I have. I’m creating a part now (in Eagle, so that takes a while) for a 7 position screw terminal.
to connect several sync busboards together (on both sides)
I’ve made 14pin shrouded headers with big oblong pads for easy soldering.
I was thinking: some alternative pins for custom sync cables / pcb connections.
and text indications of what signal is connected to what pinrow. + polarity
the pcb is 16.5 x 4.2 x 1.2 cm. is this size ok, or should it be bigger / smaller?
update 1
screwterminals , names & extra gnd pads added.
mmh. yeah, RCA is the method the other modules use right? but that is coded signal I think. so not directly connected to the pins I have here. I will ask the experts @creatorlars ? would this be useful ? or does nobody need this added?. I could make separate traces.
True. I think the 14-pin sync is 5v TTL and the RCA sync is composite sync. There still might be some fun to be had taking the TTL sync out via RCA (or BNC!) for sync “experiments”
the RCA sync signal is specially interesting for large cases with Expeditionary, Visionary or Orion series so having a RCA or even minijacks (both right angled) its interesting
Regarding the board design I would try to use thin traces to reduces resistance… maybe its irrelevant but with this small signals (1v) every mV counts, also I remember that its not recommended to use long cables… for the same reasons.
I would suggest to even add buffers. Im thinking this for large multicase setups… of course this means your bus board will stop veen passivr and will need power from the power bus… maybe Im complicating stuff
thanks for your thoughts!
I’m not sure what to do with the RCA / jack connectors now.
The thing is that I don’t have any way to test this method because I don’t have these type of modules.
I only have DIY stuff.
How do people connect the sync of Cadet modules to their VC now?
on the buffering: this sounds interesting. but maybe I will do a passive board and a buffered board later on.
the trace width is currently 0.032 inch. so 0.12 would be better? Or even smaller?
If you want RCA sync output from your Cadet 1, you need to attenuate the CSYNC signal to ~0.6Vpp (2X gain), buffer it with a video op-amp (LM6172) and drive it thru a 75 ohm output resistor and large AC coupling cap to the RCA jack.
I don’t think that’s right… resistance = resistivity × length / area, so the thicker the conductor, the more cross-sectional area there is, and therefore the smaller the resistance. That also explains why the longer the cable the more resistance.
Re:trace widths, the bigger the better! That said, I don’t off hand know what a good practical minimum would be. There are a lot of places with “trace resistance tools” or “trace resistance calculators” so if you really want to get precise, or just play around with some numbers, I’d say check those out. I know for eurorack power busboards it’s a good idea to make the ground thicker than the others, so that might make sense here too, maybe?
Re:buffers, that makes sense to me (though I’m no expert!). But I’d definitely use video rate op amps (LM6172) like Lars mentioned, since the sync signals probably wouldn’t be the cleanest through slower TL072s.
thanks for the reply. I’m no expert either, so all help is welcome
I think I’ll make make a groundplane then, and fat traces for the other signals.
I’ll check those sites too.
For the opamps. yeah, I just used the tl072 as a marker. I don’t have a Eagle part for a LM6172, but those pinouts are the same. (I’ve updated the picture though)
from the calculator site
resistance is 0.125 ohm for 0.032 inch trace
0.0801 for 0.5 inch trace
Use an all over ground plane and nice thick traces for your power rails (40mil) – or 4 layer board with internal planes for power and ground. TTL sync signal traces can be thinner (12mil) since they are not carrying current. Space them out a bit if you can to minimize crosstalk.
For buffering syncs, no need to do that for the H/V pulses sent out on the CV/Gate bus, they are already buffered. For the 14-pin TTL signals, you buffer them with 74HC14 as shown in the Cadet 1 datasheet (they are also already buffered.)
If you want to convert TTL CSYNC to an RCA video signal (black video with sync only) you need an extra buffer and driver (you need to attenuate the signal from 5V to 0.6V at the input to the buffer using a resistor divider and then drive it thru a 75 ohm resistor and AC coupling cap (47uF or so should be good.)
So you are saying that buffering the TTL sync signal is not really necessary?
Does this also take into account that if someone has multiple cases and want to connect +10 VCO’s?
I haven’t reached a point at which chaining the syncs from a single 74HC14 buffer is insufficient (even with 10+ modules attached to the 14-pin bus), but you could buffer on the busboard if you wanted to. (One sync input header, multiple sync output headers.) It certainly wouldn’t hurt anything. You’d need 5V power supply though, to do that (so 78L05 or similar.)
I may be short on time to draw up some circuits for you but if you can post screenshots here I can reply and offer advice. It’s great you’re doing this!! I really want to document all our standards and some best practice circuit examples in a formal document and open document at some point, and this is a step in that direction. It’s so exciting to me to see the community taking the initiative on projects such as this.
ok. test circuit for buffers, 5v and cv/gate bus connection. I’m unsure about that last one.
Can these nets be connected in this way?
The purpose would be to connect Castle & Cadet modules to the same busboard. maybe I have to put two jumpers in place. (?)
and the 7805, should I use a LM7805 , or a small 78l05?
I guess the 74HC14 won’t use much power, so maybe a 78l05.
This looks OK! 78L05 should be fine to just power a single IC. (so use the IC3 instead of IC2 circuit.)
For the HSync/VSync on the power bus, these are driven as LZX standard 1V signals from the sync generator. You could re-buffer them, but it would require additional opamps (so I don’t think you need to worry about it.)
