Ribbon Cable Current Limit?

[ArnieW]

I am currently designing a networked picaxe system that uses a common bus to allow simple expansion of modules. for simplicity, it seems that a 16-way IDC connector on a ribbon cable will provide this. I want this 'bus' to supply DC power as well as serial transmit/receive lines and a picaxe reset line.
I intend to common multiple conductors to increase the current capacity of the ribbon cable, but I was wondering if anyone knows what the current capacity of ribbon cable conductors is?
I have not been able to find anything helpful with Google.

thanks, Arnie

 

[hippy]

http://www.meci.com/Catalog/Category/93817f4d-dda7-4aa3-a22b-be26d8da06fe

Indicates standard 0.1" molex pin headers have a 1A max rating

http://www.leotronics.co.uk/cable.htm

Indicates it's a 1A rating for the cable they use.

This is probably not a definitive answer, but it's probably a good ballpark figure to use.

 

[ArnieW]

Thanks Hippy, I think I will be conservative with my design, but it looks feasible.

Are there any particular considerations regarding noise suppression?

cheers, Arnie

 

[womai]

To mitigate power supply noise, you have to decouple the power on both the source and the target board. A 100nF ceramic cap close to every IC on the boards, as well as a single larger electrolytic cap (maybe 100uF) anywhere on the boards is a good start (all those caps are between power and ground of course). As for the cable transmission, separate RX and TX by AT LEAST one line connected to ground (two is better) to reduce crosstalk between them. Same goes for TX/RX and the power lines (to avoid crosstalk from power into the signal lines). Keep main power and ground lines adjacent to minimize their loop inductance. Decouple power and ground with 100nF capacitors close the the cable launch points as well.

In summary a ribbon cable setup like this should work fine:

GND-PWR-GND-PWR-GND-GND-TX-GND-GND-RX-GND

If you need to stick with 10 wires, omit the left most GND.

Wolfgang

 

[ArnieW]

Thanks wolfgang,

I had planned to separate power and RX/TX using GND lines, but had not thought of decoupling RX & TX, so thanks for the tip. This one is a big project, so I'm spending a lot of time getting the design right on paper before I commit to a test board, before I make PCBs etc.

thanks to all who have helped along the way.

Arnie

 

[womai]

Just to make sure there hasn't been a misunderstanding (since you say "decoupling RX/TX&quot: I didn't mean you should put capacitors on RX and TX. That won't work. Capacitors go on PWR only. You "decouple" RX and TX by having them keep sufficient distance from the power lines and each other, and above all by inserting GND lines into that distance.

Wolfgang

 

[ArnieW]

Hi Wolfgang,

yes, I understand that - your GND - PWR - GND etc. 'schematic' of the ribbon said that clearly. I had not thought of putting a GND wire or two between the TX/RX lines, but there is no reason why this cannot be part of the design. It makes good sense to me.

thanks Again, Arnie