Post a pic that makes you LOL

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Not to mention, all those variable lead lengths would royally screw up timing on a processor running at GHz speed. Remember, 11 inches = 1 nanosec...so those longest leads will skew timing by a significant chunk of a ns...
It's worse than that. That is insulated wire, so the electric fields see the permittivity of the insulation. I'll guess insulation is Kaynar/PVDF so Er is up to about 8.5. So 1ns is potentially as short as 35mm (1-3/8").
 
Airsoft and bb guns are par for the course in rc ship combat(unmaned). The ships have damage areas that can be repaired readily. One might try the NAMBA web site for more info.
 
It's worse than that. That is insulated wire, so the electric fields see the permittivity of the insulation. I'll guess insulation is Kaynar/PVDF so Er is up to about 8.5. So 1ns is potentially as short as 35mm (1-3/8").
Umm, no. 11 inches at c is about 1 nanosecond. Higher permittivity doesn't make the signal propagation exceed c. The signal will be slowed, not accelerated, by about sqrt(8.5)*, so 1 ns is 87.5 cm (34 in).

Still, the crosstalk, parasitic capacitance, and surely other things will mess it up more than enough.

* The speed of an EM wave is 1/sqrt(µε).
 
Last edited:
Umm, no. 11 inches at c is about 1 nanosecond. Higher permittivity doesn't make the signal propagation exceed c. The signal will be slowed, not accelerated, by about sqrt(8.5)*, so 1 ns is 87.5 cm (34 in).

Still, the crosstalk, parasitic capacitance, and surely other things will mess it up more than enough.

* The speed of an EM wave is 1/sqrt(µε).
I am pretty sure I was right in that the signal is slowed by the higher Er around the conductor, so the distance travelled in 1ns is going to be less than the distance travelled in free space (can't exceed the speed of light in vacuum). By implying the distance is increased to 34" you are incorrectly applying the factor to increase the velocity. Will double-check the scaling factor though. Hold that thought...
 
Umm, no. 11 inches at c is about 1 nanosecond. Higher permittivity doesn't make the signal propagation exceed c. The signal will be slowed, not accelerated, by about sqrt(8.5)*, so 1 ns is 87.5 cm (34 in).

Still, the crosstalk, parasitic capacitance, and surely other things will mess it up more than enough.

* The speed of an EM wave is 1/sqrt(µε).

I am pretty sure I was right in that the signal is slowed by the higher Er around the conductor, so the distance travelled in 1ns is going to be less than the distance travelled in free space (can't exceed the speed of light in vacuum). By implying the distance is increased to 34" you are incorrectly applying the factor to increase the velocity. Will double-check the scaling factor though. Hold that thought...
Ok. I concede that I forgot the square root. Apologies. So 1ns with Er of 8.5 would give a length of about 103mm (4").
 
I worked in an R&D lab and if someone had asked me to do this job my response would not have been printable here. It might have included references to places that never see the sun, etc, etc. I've done my fair share of soldering leads to the pads and vias on PCB assemblies but I would never want to try this.
edit...
Surely it would have been cheaper and faster in the long run to just spin the PCB? I had a prototype board given to me that didn't work. After checking for a bit I discovered that the board had been laid out with the footprint for the processor backwards/upside down. Rather than try to blue wire it they just spun the board and all was right with the world.
One of the first boards I worked on at a new company, using a new layout tool, came out like this. With five 100+ pin FPGAs, I managed to create the footprints viewed from the bottom rather than the top. But, they were all Pin-Grid Array packages. We installed them on the back side of the board and they worked fine. If anyone asked, it was for "cooling issues". Fortunately for me, there were other changes to be done to it, so we did respin the board, but the 'backwards board' was fine for testing.

But if we had a rewire to do, there was a row of 'blue-haired ladies' that would do any silly wiring you asked for, with amazing craftsmanship.
 
I am pretty sure I was right in that the signal is slowed by the higher Er around the conductor, so the distance travelled in 1ns is going to be less than the distance travelled in free space (can't exceed the speed of light in vacuum). By implying the distance is increased to 34" you are incorrectly applying the factor to increase the velocity. Will double-check the scaling factor though. Hold that thought...
Slower: yes. Slower means less distance in a nanosecond: yes. I said more distance but I should have said more time: uhh... yes. Here I was thinking I was standing on my head and thinking you were standing on yours. Those wires, at the length they are, will add more delay than they would at c, which is what we both meant. That is, it's not really the distance for a given time that matters, but rather the time for a given length of wire, which goes up, and stupidly I said the distance goes up instead.

But I'm still pretty sure it's by the square root, approximately, and somewhat more complicated than that. Just try applying the telegrapher's equations to a rat's nest.
 
Back
Top