Blue Mountain Avionics EFIS 1
9/2/2005 13 Hours
No pictures, just a story. Today, I spent about three hours of my day trying to get the EFIS 1 working with the GX-60 that Mike brought around the other day. When we initially plugged it in, it obviously didn't work off the bat. I went home and did some reading before coming back today with the resolve to get it going. I was confident that it wouldn't be a problem too. Well, the Vertical needle for glideslope worked fine. For some reason, the CDI wasn't playing though.
I checked the outputs from the GX (except for the Nav Valid ones) and called Bob at Blue Mountain. He suggested I check the outputs for Nav valid. Turns out I had 'em reversed, so it would never trip the needles. In order for the needle to swing, the E1 MUST have a Nav Valid flag.
So, now that was all hunky dory, I figured it would work. Nope. Nothing. Well, there was something. For whatever reason, the sensor calibration screen showed Loc Good when the From?To flag was being raised by the GX! What's going on?!
Bob also suggested I put everything in an email, so here is a copy....
Start of email...
Bob,
We spoke on the phone this morning with regard to my issue of no CDI needle. I checked my
wiring thoroughly (took a couple of hours) with a multimeter, and all seems to be good. I
did find that I had the Nav Valid + and - switched, which obviously would not work, so I
switched these over and something did happen, but not what I expected.
Let me go over my installation first so everything is clear. I have an SL30 connected
using the serial connection (Serial 2. I have FADEC into Serial 1.) I have my GX-60 and my
SL-30's analogue outputs wired to the analogue 1 on the EFIS 1 through an Apollo ACU. This
is to enable me to legally fly a GPS approach using the outputs from the GX-60. Having the
SL-30 connected will also serve in case a serial line goes out for whatever reason. I
didn't really need to do this part, but I'd cut my panel for the ACU way before we figured
that we probably didn't need it, so I went ahead and used it anyway.
On the Autopilot settings, I can choose then what I want to use as the input for the
External VOR. All the external options can be either SL30, or in my case ACU (was probably
GX-60 originally).
Now then, let me tell you what happened when I corrected my Nav Valid wires.... I didn't
get a CDI needle still, however, when the GX cycled on startup, when it said testing the
From/To flag, the sensor cal screen showed Loc Good! This immediately prompted me to check
my wiring! Maybe I'd wired the From/To into the Nav Valid. Well, 'twas not the case. The
sensor cal screen shows that the GX is sending the outputs to swing the needle OK. It's
just that it's not getting the Nav Valid for some strange reason. The Vertical stuff works
perfectly!
I couldn't use the test mode for the GX to test the needles as it doesn't output a Nav
Valid or GS Valid signal when it swings the needles for whatever reason. I could use the
test mode to calibrate the CDI/VDI though. The only way to test properly was to cycle the
GX.
For the wires from my ACU, I don't have anything connected to the From/To at all, as
there's no inputs that I can see on the EFIS 1. On the phone, you mentioned that it did
have these inputs. I looked again through the manual, but found no reference. There is a
From/To input on the G3. Am I missing something here?
I'm going to post this email to the board too in case someone else has has a similar
problem and found a solution.
Thanks for your help!
Regards,
Mark.
End of email...
So, watch this space. I posted this on the discussion board on the Blue Mountain site too, so maybe someone has this figured out already.
9/11/2005 Update
I've been chatting to Bob on the phone some this week, in an effort to figure out this strange phenomenon. He couldn't understand why the EFIS 1 was getting all the info it needed, but wouldn't put out the display. I'm baffled too, so what I plan to do is this....
I'm going to try to find an avionics test set to see how the EFIS 1
reacts to the SL-30 analogue outputs. If this works as desired, then the GX-60 must be
doing something funny with it's outputs that we don't know about. I think this will prompt
a call to Garmin to find out why.
If I cannot find a resolution to the problem, I'll be doing some cheating, and applying a
small voltage to the NAV Valid flag into the EFIS all the time. This will then make it
think everything is great, and force the CDI display. I can't really see too much of an
issue doing this for the outputs from the GPS, but I'm hoping that the SL-30 behaves, and
I don't need the cheat for the SL's outputs. We'll see how it goes.
If anybody out there can see an issue with fooling the E1 into displaying the CDI, let me
know! Thanks!
9/24/2005 8.5 Hours
Today, Carl came around from USA Jet with an avionics test set. We ran the radio stack through it's paces and watched what the EFIS 1 did. It behaved pretty well! It seemed a little too sensitive to the CDI outputs from the radio, so I need to check that.
The main reason for having him come over was to help troubleshoot the CDI needle for the GPS. Well, we discovered that the SL-30 was working as expected in digital mode. We needed to switch it to resolver mode in order to check the EFIS 1's analogue inputs (which I wasn't expecting to have to do) but that worked perfectly. So, CDI needle swings OK, and the Glideslope swings OK. They're off the scale when Carl set half deflection, but that should be an easy fix. That means then that the EFIS is functioning correctly. My wiring from the ACU is correct. The NAV Valid Flag is working from the radio. Sweet. Now to figure out why it doesn't do it with the GPS.
First I decided to try hooking up the GPS antenna to the GX-60 and wheel it outside. Ah ha! Now we have some valid Nav info, the CDI needle comes up! Perfect! I guess I should've tried this in the first place eh? Well, I'm happy today after all that. I wonder why it doesn't swing the needles during the IFR test at start up?
So, the way I'll be using my set up will be to have the serial interface as primary for the SL-30 to get the OBS functionality, radio frequency switching etc. and then have analogue as primary for the external GPS. If my serial connection goes awry during a flight sometime, I can switch the SL-30 to resolver mode quite easily and drive the EFIS with that. Perfect.
1/25/2006 2 Hours
There was a service notice a while ago on my version of the Blue Mountain autopilot servos. The clutch mounting plates were coming loose as they wore, so I have some new ones to put in. I figured a good time to do it was now since I had to remove the autopilot servo from the wing after I noticed that I'd run the cable to the wing tip instead of the wing root. Duh! At least I proved it was possible to remove the servo with the wing skin in place! It was harder getting it out of the wing access hole than it was to get it through the lightening hole in the rib, because of the orientation of it, and the upper wing skin being in the way, but it came out OK.
Here's the clutch end of the servo opened up. I need to remove this clutch and put in the new mounting plates. Not too hard. A bit fiddly to get access to the set screw in there, but not that bad.
Here are the new mounting plates. They use two set screws instead of one. A bit of added security. I still haven't completed this task. I'll wrap it up on Sunday.
5/16/2006 Update
On the way to the hangar, I stopped by a motorcycle store to buy the smallest gell cell battery they had. The other night, I'd tested the system to see if I could hit the starter motor and not kill the EFIS. Well, it didn't do as I'd hoped. I was hoping that there'd be enough in the battery to sustain the EFIS during start up, with the PC680 battery and the Sky-Tec starter motor that supposedly draws less current than all the others. I hit the starter switch for a nanosecond, and the E1 shut down, and started rebooting. Oh well. It had been warned about in the past. I was just hanging on to the slim chance that it wouldn't happen to me.
I got a baby Yuasa battery. It's pretty small. I'll take a piccy of it and post it tomorrow. I got a gell cell one so that I can orient it in any position that takes my fancy in the plane. Plus, it won't leak acid all over the place if I do a loop or roll and accidently pull -ve G's! All I have to do now is find somewhere to put the thing! I ordered a shottky diode from Allied Electronics to stop the starter from sucking the life out of the battery at start up. This is enough to keep the EFIS going through the start up process. Important to me, as my engine gauges are on it! Yikes!
6/11/2006 5 hours
Most of my day today was spent messing about with the EFIS 1, and getting all my engine gauges working as they should. I needed to get the ammeter working properly, which meant moving the input from one of the fuel senders to a different pin. Reason for this is that there are only three 10v voltage inputs on the E1, and I have them all used up! I have the fuel senders into two of them, and the voltmeter into the other.
There are a whole bunch of resistive inputs in the E1, that from the factory come as voltage inputs. To make them resistive, you have to install a jumper. These only accept 1.67v though, so in order to make it all work out some thinking was in order!
The current sensor from Blue Mountain outputs 5v when the current is close to zero amps, and it comes with a calibration table to input into the sensor map to calibrate it correctly. This means that it would be best for me to use one of the 10v inputs, otherwise I'd have to spend a bunch of time messing with the calibration. Something I didn't really feel like doing.
The answer was to move one of the fuel sender wires to the 1.67v input and then use a 10v input for the current sensor. All would be good. My fuel senders send out up to 5v too though, but I could easily make a voltage divider to bring this down to an acceptable level. That's basically what I did. It's a lot easier to calibrate a fuel gauge than mess around calibrating my ammeter with a multimeter!
I also loaded the calibration array for the Floscan fuel flow meter. I hadn't done this yet, so I was getting all sorts of wacky numbers for fuel flow! I use a 201B sensor, and there's a pre-loaded cal array in the E1 just for that.
The E1 used to display OAT on the ADI part of the screen by the true airspeed number. I notice that this has now gone, so I decided to add that to the engine gauge page.
I ran the plane a couple of times today too, to confirm everything was good. Fuel flows around 2gph while sitting on the ground, and the ammeter works fine now too. We now have 0.8 hours on the engine. Sweet.
I'm almost ready to put that top forward skin on! I need to pluck up some courage though first!
I also put the taxi light in. I didn't bother putting it on the wings page, as it was just a trivial 5 minute thing, and it's the same as the landing light. I just figured I'd log it here for good measure.
While I was at Oshkosh, I chatted with Blue Mountain and they suggested I send my unit in for a preflight check out. During this check out, they'll update me to the latest software revision level, check all the sensors for correct operation and make sure everything is calibrated correctly. I removed the unit today ready for shipping it off. I'll give BMA a call tomorrow to check in with them to get the 'pre-flight' done. Sweet.