MPPalmer(at)aol.com
Guest
|
 Posted: Tue Jun 17, 2008 3:13 pm Post subject: More on Tom's RFI problem |
 |
|
Hi All (and Tom):
Sorry for the delay. I find it hard to read ever since an angry Mormon gal punched me in the eye on Christmas Eve 2005 while I was doing some Christian missionary work. Her punch caused a torn and detached retina. I had emergency surgery, but my dominant eye hasn't been the same since, and for now, I'm reduced to reading with one eye. (That was felony assault. I called 911. She fled the scene. I was the only one bloody, but the cop said it was "my fault" for being there! So much for the 1st Amendment in the United States.)
It's not clear from Tom's description whether they've solved any of the problems yet. If not, I would focus on the Conducted interference in the Tru Trak first. If you can kill off Conducted there, it might ameliorate some of the radiated problem in the rest of the panel. So put the dummy load back on.
If the offending transmitter is fairly new (like, the last 10 or 15 years) and a TSO'd unit, it "ought" to be pretty good with internal filtering. When I worked for Cessna Avionics, I did some of the TSO testing for radios in design and the TSO is a pretty robust test. (Even better if you have to meet German specs!) So you shouldn't need to work too much on the power input to the transmitter, although a ferrite bead or two can't hurt, since we know some RF is getting out the radio.
There are these "split" type squares that Radio Shack used to sell and I've gathered some cylindrical ones from computer products I've accumulated over the years. Some USB cables on better equipment have them or sometimes the power cord on a laptop computer will have them. Lately, they've been molded in (better) and don't come off. But some have two plastic clips or tangs that you can pop and the bead opens like a clam shell.
For an initial test, you can pop one of these on your existing power wires to the COMM (notice the plural... our radio shop ran two 18 g wires for power (and ground) to the COMM. If the Tru Trak gets better or the problem goes away, then you know you're on the right track. (The split type ferrite, being split, will never be as good as a continuos piece of ferrite. But they make for good non-destructive troubleshooting of RF.)
Similarly, and perhaps more important, put some of the computer ferrite thingies on the power of the Tru Trak. (Is the Tru Trak TSO'd?) Even though you said Radiated wasn't a problem with the autopilot, I'd also put some beads near the unit on the wires to the servo(s). It may be that they're re-radiating some RF and causing problems with the other radios, which will manifest later after you take the dummy load off.
Now, the Radiated is harder to solve. It's a lot of guessing as to what could be the culprit. Since the dummy load test eliminated the interference in the rest of the radios, the question is, how is the RF getting to all the other radios. (What are those radios, BTW? You said "everything in our panel.") If it really is ALL the other radios, then that ought to be a clue. If all the other radios are affected, it's unlikely they all have the same vulnerability to RF, so we're looking for a vulnerability common to all.
What happens if you take a large sheet of aluminum (or some other conducting metal) and hold it between the radio stack and the transmitting antenna as a giant shield? That might give you some information for trouble shooting. Move it around various parts of the panel - even block the batteries. Any changes?
TSO's not withstanding, I've heard there are problems with some COMMs interfering with some GPS's. But that doesn't sound like the small type of problem you're having.
Now, I don't know you're electrical layout. You talked about two batteries. Do you have any diodes in that loop? Being nonlinear devices, diodes are notorious for picking up RF and re-radiating it at different frequencies. Diodes across starter solenoids and master contactors are okay, but, say, if you have a diode in the power lead of one battery to isolate one battery from another, that would be bad news, RF wise.
This will sound confusing, but I would try bypassing the power on all your electrical bus to your avionics and power everything up with a single battery close by the radios. While that sounds like I'm testing for conducted interference, I'm testing to find out if RF is being radiated into any outside power wires and then conducted into the radios.
I assume that you, like us, are using unshielded single wire for your transmit switches, flap indicator(s?), the trim connection from the stick to the relays, etc.? Not that this has been a problem for us (that I know of) but if I had to do it again, I would use some kind of thin audio shielded coax for any control wires coming out of the stack. In this day and age, that would include the GPS annunciator thingie the FAA mandates, etc. For completeness, you might want to try disconnecting all those extraneous wires as they make great antennas (especially if they just happen to be at the right length) to see if any of them are picking up the RF and then conducting it into the stack.
I once flew to Chicago to help a daring Glasair builder fight RFI on an intercom system I built and offered through S/H years ago. (A not very professionally crafted intercom on my part. Similar to the stuff Jim Weir prototypes and shows in KITPLANES.) I spent two days trying to isolate the problem and in the end, the solution was to run every wire through an extra outer shield (braid), in and out of the intercom. (Making, for example, the audio cables "tri-ax.")
Lastly, how is your ELT plumbed into your panel? Our ACK used to be sensitive to RF and certain freq's on COMM2 would set it off. Part of the problem was that silly telephone cable they use from the transmitter to the remote head. Knowing what I know now, I would have slipped that in a braid to shield it at the panel. I suspect the ELT would pick up COMM stuff with it's antenna and send it to the panel via the annunciator's telephone cable. (I don't know if the ELT TSO covered annunciators back then, so there may not have been testing for that.)
Okay? Short of seeing your plane for myself and getting hands on (maybe if I get my medical back when my eye is fixed?), I'm running out of ideas. This RFI stuff is kind of an art and you have to be creative in guessing about how the RF is getting out and getting in. The power output of an antenna is a "one over D squared" relationship (in the far field), so the closer you get to the antenna, the much larger the RF. (I think in the near field, the power relationship is even greater. But my text books are buried away.) It always amazed me during TSO testing how tightening the screws on the case of a radio could make the difference between passing and failing a radiated test. Imagine there's a light bulb in the radio and you're in a dark room. Any light leaking out is bad. And, like an old film camera, any light that can get into another box is bad. That's the level of the problem you're tying to solve.
Let us all know what solves the problem.
Mike Palmer <><
+++++++++++++++ ORIGINAL MESSAGE +++++++++++++++++++++++++++++
You responded to an email of mine on this topic several months ago about comm RFI
EMI getting into everything in our panel and you suggested to sort it out to
put a dummy load on the comm pigtail and key up and see what happened... we
did that and all the trouble went away except the Tru Trak AP still had a heading
change. So we concluded that we have both Radiated and conducted RF troubles.
We did VSWR testing on our old antennas and they weren't too bad but changed
them anyway from the Glasair Aviation copper foil dipole to the Bob Archer
006 with a well made connection and a proper install. Recall Bob Archer does not
advocate the use of a balun. The VSWR didn't improve much... about 2-2.5 The
problem persisted. We also found 1 very leaky B&C connector and fixed it, we
routed the comm coax from the center of the plane to the fuselage side wall.
We tried the AAE dipole with the built in balun and it seemed better than the
Archer 006. We had a nice long pc of RG 58 coax that
I was able to walk around the hangar with ranging 20-30 feet from the Glasair
and keyed up and things were alot better at a distance and when we keyed up close
to the plane the problems were worse but better than the Archer 006 with no
balun.
Our grounding seems good with a few exceptions. All the avionics and all panel
electrics... lights, pitot heat, strobes etc are brought out to one common ground
10g wire that then goes to the copper strip avionics ground under the panel
and that strip is in turn grounded to the engine block via 4g wire. Recall from
my email a few months back that the panel is profesionally built and that
company set up the 10 ground for us to run to the av ground bus. The 2 batteries
negitives are brought together on 2 g and then through the center console
through the fire wall to the engine block near were the avionics ground bus attaches
to the block. There is good ground continuity from the avionics ground
bus back to the batteries with resistance in the 300 mohm range. The exceptions
to the grounding are that the ray allan trim and flap indicators, the trim connection
from the stick to the relays and then to the trim motor and the flap
connection from the flap momentary DPDT to the flap
motor may all or partly be doubly grounded. These connections were all provided
for by the panel maker and therefore the ground accounted for through the 10g
avionics ground wire. We ran into trouble using his connections because the
wiring for them wasn't big enough so I think they are powered and grounded separetly
but still connected to some extent via the panel makers original wiring.
So we may have a ground loop somewhere. So my thought is to disconnect items from
the avionics ground bus in an organised way and key up and see if things improve
before we resort to chokes etc.
**************
Gas prices getting you down? Search AOL Autos for fuel-efficient used cars.
(http://autos.aol.com/used?ncid=aolaut00050000000007) [quote][b]
| | - The Matronics Glasair-List Email Forum - | | | Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:
http://www.matronics.com/Navigator?Glasair-List |
|
|
|
Get Email Distribution Too!
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
