USB and electrical fields

[dc8flightdeck]

After adding solenoids, motors, and 24v DC lights to my Phantom sim it became unstable. The USB hubs would go offline anytime I toggled a switch.
Sometime it even caused the computer to reboot. I thought maybe the issue was the common ground going through the aircraft chassis, but the
canopy switch was completely isolated and not physically wired to anything connected to the computer. After I confirmed there wasnt a short I
looked elswhere for the problem.

It turns out my electrical gear was causing electrical fields that were disrupting the USB signals. The computer is 20 feet from the sim so I use long
cat 5 to extend the usb to 3 hubs in the sim. I carefully seperated the USB from proximity to any power cords, and what a difference that made!
So far the sim is running with no crashing.

This USB problem has become an issue on several of my sim projects where I run real 24v aircraft parts such as a real autopilot panel on a 737-200
(it uses solenoids). When I turn off the auto pilot, the solonoids are triggered via FSUIPC to move the real 737 auto pilot switches to off, and the
computer USB connections go offline. Ive the same problem with multiple types of cards.

Has anybody else experienced problems like this?

All of my projects use real sims or real cockpits so they all have chassis grounds due to the nature of real aircraft wiring.



Justin

[jonesthesoftware]

Hi Justin
this is an extract from WIKIPEDIA( hope I'm breaking any rules here) but it explains why you're having problems. To cut through the jargon maximum length of USB cable is 5 meters. DOn't be fooled by companies selling extra long cables, they might work for some devices like printers but not for what you are doing. Use a 5 metre cable to a powered hub then another 5 meters from that might help your problems
regards
geoff

USB 2.0 provides for a maximum cable length of 5 meters for devices running at Hi Speed (480 Mb/s). The primary reason for this limit is the maximum allowed round-trip delay of about 1.5 μs. If USB host commands are unanswered by the USB device within the allowed time, the host considers the command lost. When adding USB device response time, delays from the maximum number of hubs added to the delays from connecting cables, the maximum acceptable delay per cable amounts to 26 ns.[35] The USB 2.0 specification requires cable delay to be less than 5.2 ns per meter (192,000 km/s, which is close to the maximum achievable transmission speed for standard copper wire)

[dc8flightdeck]

Thanks, thats interesting info.

In the 737 the USB cables are about 5 feet, so I dont think length is a problem. Its only when power is toggled at the solenoids that the USB shuts off. Dont use them and the 737 runs fine.

On the B-52 sim it works like this, the open cockpits relay card is connected to real 24v dc landing gear indicators. The relay card's usb is connected to a networked laptop. The main pc has usb lines connected to a pot, switch input, and multi controller card. when I toggle the landing gear the main pc usb goes offline, but the laptop usb is ok. Turn off the 24v power supply and the usb on the main pc no longer goes off when landing gear is toggled.

My experience with long usb cords is that they dont work at all, thats why I switched to cat5 with usb adaptors. That worked ok in the Phantom sim until I added more electrical emitting equipment inside the cockpit. The cat5 cables ran in a bundle together with the power supply lines and separating them seemed to do the trick for that sim. It ran perfectly last night.

[jonesthesoftware]

Hi Justin
from those results it seems the fault is simply that of induced currents, i.e, power cables when switched ON/OFF are inducing currents in your lower powered data cables like the USB. Only answer is to separate the power cables from the data cables and input cables such as potentiometer and digital input cables. Normal recommendation in industry is a minimum space of 6 inches spacing where cables are parallel. Other thing is use shielded cables BUT the shield must only be connected to ground at one end. If you have any shielded cables with both ends of the shield connected to ground this causes a ground loop for induced currents and is in fact worse than no shielding. One final recommendation ...the relays that switch the solenoids should have "shunt" diodes connected across the load to absorb the reverse current pulse which is generated by the switching of a solenoid. This also prevents burn out of the relay contacts
regards
geoff

[dc8flightdeck]

Geoff,
Friday we placed diodes at the landing gear indicators in the B-52. That made a huge difference. The attitude indicator still jumps a bit when the gear is toggled, but it didnt cause anymore problems with the USB.

Thanks!
Justin

[verticallimit]

I build my simulator in a caravan, this gave many problems with noise on the USB port until all computers monitors and power supplies were grounded to the earth.

[737NUT]

Sounds like a bad Ground loop issue. I had similar strange behavior in the past and it was repaired by grounding pc cases to the cockpit frame and even some of the usb outer connector shells to gnd.

Rob

[jonesthesoftware]

Hi Justin
what about the solenoids did you put diodes across these, as these are the worst cause of induced voltages/currents?
kind regarda
geoff

[dc8flightdeck]

Geoff,
The 737 is away for a while, Ill have try diodes on it in a few months when its back in the hanger.

On the B-52 the problem was with the six barbor pole landing gear indicators, which function similar to a solonoid. The B-52 sim doesnt specificaly have solonoids, the problem was caused by the barbor poles.