Thursday, December 30, 2010

N951CC is flying again!

After 4 months of downtime, N951CC is finally flying again! Although there were some airplane fixes such as an exhaust leak and a fuel tank leak, it was really all the extras and new toys that made it take so long. Specifically, I had to wait over a month for a hose fitting for my backup vacuum pump. That kind of lead time can really extend a project.

I've been trying to do as much ground testing as possible along the way to ensure everything was working properly. The ground testing was going well and only found 1 issue, a burnt out low vacuum warning light. This was a new light that was dead on arrival, but fairly easy to replace.

When we were finally ready to run the engine, it ran pretty rough, but my mechanic quickly noticed that the right cylinders weren't firing. After working on that for awhile and checking different items, I found a slit in an induction tube that was leaking air into that bank of cylinders. It wasn't too long until that was replaced and the engine was running on all 6 cylinders.

I was able to take it for a test flight the next day after the fuel tank top panel was sealed back up. I came back with a new list of squawks. I had a hot mic, the fuel flow gauge was reading 0.0 GPH, the digital tach was reading high and the digital manifold pressure gauge was reading high. I was able to turn off the aircraft radios and use my portable in order to continue the test flight without the hot mic.

After inspecting the audio wiring, I was able to get the mic to disconnect. I'm not sure exactly what it was, which makes me worry that it could happen again, but I think it's the co-pilot push to talk wiring because it is in really bad shape. I've got a replacement cord on order. On the second test flight, I again noticed the high readings on the digital tach and manifold pressure gauge. I also found that I could not transmit on comm2. Comm1 worked completely, and I could listen and key the second comm, but not transmit.

I swapped comm1 and comm2 radios and the problem was still on comm2. I was down to thinking it was a bad audio panel or connection between the audio panel and comm2. A lot of wiring was moved/retied/pulled/pushed/etc behind the panel during the downtime so a bad wire between the audio panel and comm2 was a real possibility. I was really hoping that wasn't the problem because that would be a real pain to fix. I pulled out my audio panel while my mechanic was looking for a similar panel to swap for testing. While I was waiting, I decided to see if I could transmit from the co-pilot side. I had tested from the pilot and the factory hand mic, but not the co-pilot. While going through the logic of the connections, I was thinking that the co-pilot connections couldn't be the cause, but I decided to try while I was waiting anyways. Sure enough I was able to transmit on comm2 from the co-pilot seat! So I retested from the pilot and factory hand mic and they both now worked. Apparently the audio panel did not seat properly when I reinstalled it last time and it went in good this time. Comm2 problem solved!

Now I was down to high readings from the digital EI probes. I grabbed my mechanic's optical tach and an optical tach I have for model airplanes. The optical tachs seem to be right on, they read exactly correct when reading a light bulb and are identical between the tachs. I fired up the airplane and compared the factory mechanical tach, the 2 opticals, and the EI tach. Here are the results:

Factory / Optical 1 / Optical 2 / EI
1000 / 1080 / 1080 / 1100
1500 /1560-1590 / 1560-1590 / 1590
2000 / 2040-2070/ 2040-2070/ 2090

I'm not sure what to think of the optical tachs jumping back and forth between two values. They both jumped exactly the same way too! The documentation from EI says the RPM accuracy is 1% or better.

EI error assuming optical tachs are truth with 0% error
Based on average / Based on max
+1.85% / +1.85% @ 1000 RPM
+0.95% / +0.00% @ 1500 RPM
+1.70% / +0.97% @ 2000 RPM

So although the EI RPM probe is out of spec, it's not THAT bad. I figured my factory tach would read low, that's pretty common, but I expected the EI digital probes to be a lot closer to the optical tachs.

Next I was on to testing the manifold pressure gauge. First, we went around and took samples of MP readings on airplanes around the airport. We found 2 other airplanes that read the same as mine, 1 a half inch higher, 1 a half in lower and 1 an inch and a half lower! Obviously this was not going to help me verify the accuracy of the EI probe. After trying to figure out a way to get an accurate pressure reading to compare against, I found that I could correct the AWOS altimeter setting for altitude to obtain an absolute pressure reading. I also called EI for their suggestion and they emailed a troubleshooting guide that outlined the same calculation. Armed with the math, I calculated the absolute pressure and compared against the factory gauge and EI probe. The AWOS altimeter setting was 29.72 with a field elevation of 2457 feet so the calculated absolute pressure was 27.263 inches. With the space between 1 inch ticks on my manifold pressure gauge only about a half centimeter apart, the best guess of the factory reading was 27.3 inches. The EI probe was reading 27.9 inches, significantly higher than the calculated pressure. EI states the accuracy of the MP probe is 2%, but assuming the calculated pressure is 100% accurate, then the EI gauge is +2.2%. From the SuperAWOS website, there is a reference that states that the system meets accuracy requirements for Advisory Circular (AC) 150/5220·16B. I couldn't find a copy of this AC online since it was cancelled by 150/5220·16C, but 16C states a 0.01 inch root mean square error or 0.02 maximum error in the redundant pressure system. This equates to an AWOS error of 0.037%, which is an insignificant factor. Again, the EI sensor is out of spec on the high side, but not by much (0.2 percent above the spec).

I don't think it's worth sending the sensors back for new ones. I may just compare pressures and RPMs over time and develop calibration sheets. I expected the digital probes to be more accurate, but every measurement has it's limitations.

The EI fuel flow is still reading 0 and all of the ground testing looks good. I discussed my testing with EI and the confirmed the steps I took and also believe my wiring is correct and the error is in the transducer. After our flight on Saturday, we'll disconnect the transducer and see if there is anything blocking the turbine, but my anticipation is that we'll have to send the transducer out for a replacement.

Overall, I'm pretty happy that with all of the work completed, there has not been any lingering squawks besides the fuel flow, but we have a plan for that.