Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Several ground planes on a one-sided PCB. How bad?

Status
Not open for further replies.

mik3ca

Member
Lately, I've been making a habit to myself to make sure all grounds of all IC's connect to one common ground plane for optimal circuit operation. Now I find myself in a situation in which breaking the ground plane in two and connecting them with a 2cm jumper will allow my board to auto-route 100% in Eagle.

My question is just how bad is it to do this where only two IC pins that require connection to ground are connected to ground through a 2cm jumper wire instead of a ground plane?

And would it be equivalent to a ground plane in a digital world if I used like maybe 3 or 4 unshielded jumper wires in parallel and soldered together?
 
I included a perfect snapshot of what I'm trying to achieve. Basically I'm making an 8051 series programmer using 74HC574's to hold bytes of data and 74HC245 to send data to the writable port. Shown here are the two IC's connected to P0 and P2 of an 8051 microcontroller. I want to feed the data lines and in order to do so, they have to cross the path that the ground wire (colored in cyan) is blocking. For me to achieve this on a single sided board, I need to disconnect that wire from the point where the capacitor and 74HC574's ground pin connect and connect that point to ground through a jumper wire to a ground plane. Let's assume the outer black is the ground plane.

Resistor R5 is a jumper wire (0 ohm resistor) that connects the 74HC245 VCC to +ve rail.

So could I get away with a zero ohm resistor as a jumper wire, or should I put multiple 0-ohm resistors in parallel? I want to make the circuit still work.


circuit.png
 
It depends entirely on the circuit. If it is a signal amplifier for a microwave radar, then no. If it is a low speed digital circuit, then probably yes.

What is the highest clock speed, signal frequency, etc? What are the decoupling capacitors (value, construction)? Does the jumper have to be as long as a 1/4 W resistor, or can it be something shorter and fatter, like a piece of #18 solid that jumps a gap of only 0.25"?

ak
 
It would be the speed of a standard parallel port (Probably 115Khz?) but then I may be using a 3Mhz crystal. worst case. 20Mhz.

The decoupling caps are 47nF ceramic and the jumper wire has to be at least 12mil times 16 because it needs to go over all of the bit traces in the data.
 
It looks like you should be able to do what you want with a different layout. In any case, with this circuit, neither a ground plane, nor a few jumper wires would make any difference. I'd be WAY more concerned about your trace clearances between those through holes...

Change that layout! And don't use the auto-router for your final routing!!!
 
I'm aiming for 12mil clearance between traces as opposed to the standard 8-10mil. thats why the resistor symbol pointing upward appears to be 3 cm instead of 1.5 cm from hole to hole.
 
Why not knit together the two ground planes that you see as essential with vias to the bottom? Without seeing your whole board, I am surprised that you cannot eliminate that jumper with a small adjustment to the placement of parts.
 
Why not knit together the two ground planes that you see as essential with vias to the bottom? Without seeing your whole board, I am surprised that you cannot eliminate that jumper with a small adjustment to the placement of parts.
The clue is in the thread title. :)

Mike.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top