Need SPICE model for breadboard characteristics

[BobW]

After breadboarding a circuit that worked really well, I made a PC board for it, mounted all the components, tested it, and it doesn't work. I thought maybe my original breadboard was a fluke, so I rebuilt the breadboard with a new set of components (just in case the originals were off spec in a lucky way), and it works just like the original. So, apparently I need to add a component to the PC board that incorporates all of the stray capacitances, inductances and contact potentials of one of those white plastic breadboard things. I'm sure I'm not the first person to need one of these, so I thought someone might have a SPICE model. Yes?

 

[alec_t]

I don't think a universal breadboard model would be practical. Consider just one breadboard variant. For each connection point, of which there are hundreds, I think you'd need at least 8 C, 8 R and 8 L values. to account for parasitic interaction with its nearest neighbours.

 

[BobW]

Hi Alec, It was really just a rhetorical question borne out of frustration, after a couple of hours verifying the circuit, and not finding any errors, then another couple of hours with some cheap liquor. The funny thing is that this is the first time I've designed and built a PCB where I've had this constant doubt in the back of my mind about something not being quite right. And sure enough, there was a problem. The weird thing is that it's not a high frequency circuit at all; it's all DC. Anyway, it'll probably end up being some really simple goof up. Need to get some sleep, then it will all make sense in the morning.

 

[ericgibbs]

hi Bob,
Post a diagram or photo of your PCB. front/back.
E

 

[MikeMl]

I have had many examples of circuits that could not be made to work (or worked poorly; overshoot, oscillation) on a breadboard that work much better on a proper copper ground plane...

 

[kubeek]

Also post your schematic.

 

[MrAl]

Hi,

Also post your board design

A simple mistake in the board design and/or wiring could account for the bad circuit. A DC circuit will usually work better on a real board because there are no loose connections and lower connection resistance. Also, a DC circuit only depends on resistances, not capacitance and inductance, so the only difference would likely be that the solderless breadboard had a higher resistance somewhere that allowed the circuit to work better. It is very doubtful this would happen, but i guess it could.

If you have the board design then post that with the schematic, or just go over the board design again. If you wired it point to point, then go over every connection. You can also test point to point with an Ohm meter to make sure all the connections are right.

 

[crutschow]

I suspect you have a wiring error in your PCB.

 

[bountyhunter]

If it's not a wiring error, check for a ground problem. Did you use CAD for layout or did you lay it out yourself? I have had CAD layouts where the computer isolated things because they were not "labeled" correctly. If you post it, we can figure it out.

 

[BobW]

I used a generic CAD drafting program, not a PCB CAD program. So, I had to lay everything out myself, which means there are lots of opportunities for errors, especially since I've been down with a cold the past few days, my concentration hasn't been very good. I'm going to have another go at tracing it out this evening. From its behaviour, I think I know the part of the circuit where the problem is. If I still can't figure it out, I'll post the layout and schematic. Since I laid it out myself, I don't have component locations shown, and will have to add them in order for it to make any sense.