Category Archives: Engine Installation

Oil Cooler Replacement

I mentioned in the last post that after getting the wheel pants on Betty, the airplane was ALMOST ready to fly. The reason it wasn’t fully ready was during the wheel pant installation I removed the cowl. When I removed the cowl, I found this floating around inside the bottom cowl.


When I initially found it, I couldn’t figure out where it was from. So I set it aside, and knew that I was going to have to figure out where it was from. After thinking about it on and off for a few days, the type of aluminum and shape reminded me of the oil cooler attach flanges. Sure enough, the forward upper oil cooler flange had cracked through.


It certainly didn’t do all of that in the last flight. So this problem serves as a reminder to spend more time and be more thorough inspecting. With the forward flange cracked, the oil cooler found a new “home.” The new place the oil cooler wanted to be was about 1/8″ off the top forward mounting location. This said to us that there was some slight misalignment in the mounting bracketry  for the oil cooler. This placed the flanges under pre-load. The flanges are very soft aluminum and when under pre-load are prone to cracking.

After consulting with Greg, Pacific Oil Coolers, and a few other folks, the plan for the fix was to eliminate the pre-load and also to add cross tube supports between then flanges of the oil cooler. These cross tubes make each pair of flanges a unit and will prevent the flanges from “working” under vibration and engine operation. To eliminate the pre-load we added a spacer in the mounting brackets that took up the 1/8″ gap. The spacer is made out of baffle seal material (silicone) and has some give. Hopefully this will also provide some vibration isolation.

With Oshkosh coming up quickly, I decided to buy a new oil cooler to expedite Betty’s return to the air. Then I would send in the broken one and have Pacific Oil Coolers repair the flange. This way I have a spare, and if the fix proves successful, I can either use the repaired oil cooler on my next airplane, or sell it. Then again, having spare parts is always nice too… I have to note that Wayne at Pacific Oil Cooler really treated me well for all of this. After all, the problem was my fault, but they went out of their way. Terrific customer service.




The oil cooler installation will be a watch item over now. It will require more frequent inspection to verify that we have appropriately fixed the misalignment and that the flanges are no longer prone to cracking.


She’s Alive!

Well, the title pretty much says it all. We were able to get the Lycoming pre-oiling procedure and first engine start in today. Oh, and second, and third. It is getting to be about bed time for me, so I will give some more details another time (probably when I get all the videos edited), but thought everyone might like to  know.

The engine started right up, in a few blades of the prop. Seemed to run well, though our run time was limited by keeping the CHTs from overheating. All engine indications seemed to be reasonable except for maybe the fuel flow. It was showing a fuel flow before engine start. That may be a ground or programming issue. So we will have to look into that. But most of the instrumentation appears to check out. The alternator is charging. I did do a “mag check” with the electronic ignitions and the drops were larger than I expected, but it still ran smooth. We will see if that improves as the engine clears out the preservative oils etc. I have some data from the G3X data logging that I can look at as well.  But without further ado, here is the iPhone video of the first engine start.

Here are some screen shots of the EFIS during the engine run.

830 441


Oh, and we got a few kills already too… I couldn’t believe how many bugs we had to clean off the prop.


Victory! Don’t think we will be needing to run the engine again until we get to the point of taxi-testing.


Since engine stuff is pretty much wrapped up for now, or at least until we get to the airport, we decided it was time to put the wings on so we can whittle away at the remainder of the airworthiness items remaining before we move to the hangar.


Pump it up

I woke up early (for me) today to go to the UPS sorting facility to pick up my new fuel pump. I had it in hand just before 0900. I also had to stop at the local airport to pick up a couple fittings. Greg beat me to the Garaggio, and when I got home we mounted the pump and fabricated the remaining line we needed. Then it was connecting two wires and we were in business. 

After we fabricated the remaining line, we realized that we should have made it just a touch longer because it currently interferes with the left arm rest. No big deal. That will be a 10 minute job to remake that line adding an inch to each side. 

Since I was running short on time before I had to leave for a work trip, we decided to do a quick fuel flow test and also to pressurize the system. When we turned the fuel pump on the fuel pressure went right up to 27.4 PSI and held. After we turned the pump off it bled down slowly as you would expect. This photo is after about 5 minutes of the pump being off and the system bleeding down. 

Then we did the fuel flow test. We got more than five pounds a minute out of both tanks. That is over 250% of the maximum fuel flow the engine should see and well above the 125% minimum the FAA wants to see. So Greg and I were happy with the results. We ran out of time for the day, but now we are back on track to do an engine start next. 

All dressed up…

Well, today we were planning on doing the first engine start. We set that goal about 3 weeks ago, and Greg, Kevin, and I have been working very hard to make that happen. And we were very close. In fact, if it weren’t for the failure of the electric fuel pump, we would have ran the beast.

The pump failed in kind of an interesting way. The first time we ran the pump this morning, it worked fine. Then it failed. It still made noises like it was pumping fuel, it drew a reasonable amount of current, it just wouldn’t flow fuel after the first usage this morning. We thought maybe the motor seized, but usually that would draw a high current and maybe pop the circuit breaker. It just really didn’t make too much sense to us, especially since the fuel pump was working well 2 days ago and once this morning. The pump only had a grand total of about 5 minutes of time in operation. So we spent a lot of time trouble shooting to make sure it was the pump. And it certainly is. (We tried replicating the fuel flow test from 2 days ago, and it was clearly not pumping.)

My fuel pump is made by Andair, which is a company based in the UK. I chose it because it is a very nicely engineered pump, but mostly because it was the smallest pump I could find. The location where I have it mounted, under the pilot thigh support next to the fuel selector, depended on its small size. Well the issue with using this pump became clear today. Any calls to the manufacturer for support would have to happen early in the morning because of the time difference, as well as on a long distance call out of the country. Also, it seems as though none of the suppliers in the US stock the pump. They order them from the UK once they have a purchase, then send them to the customer. So at best, you are looking a week to get a replacement if it goes bad (again).

As Greg said today, “I can just see you sitting in Left Armpit, North Dakota with a failed electric fuel pump. You would have 7 days to enjoy the local sights while you wait for a fuel pump to arrive from the old country.” He is right. So we called up AirFlow Performance. By the way, Don and Kyle and the whole AirFlow Performance team are great. Very knowledgeable, always willing to take time to educate their customers, and even helped out giving last ditch ideas to see if a competitors pump could be salvaged. Anyways, Kyle told me that they had re-designed their fuel pump to make it smaller. He sent me the drawings of the pump, and it turns out with some re-arrangement we can fit it in the allotted space.

To do this we are re-locating the fuel filter to the right side, and putting the new fuel pump in the location that used to be the filter and fuel pump. It requires making up some new hard lines to make it all work. The lines are longer and have more bends than the previous setup, but judging by the past fuel flow test, the extra bends shouldn’t be overly restrictive. Greg and I spent the rest of the afternoon working on making the modifications that will make this possible. We made up new hard lines to move the filter. We need to make one more hard line that attaches to the pump, but need the pump to do that accurately. Kevin and I also ran two new wires for the new pump as it draws more current and therefore needed larger wires.

Kyle overnighted me the fuel pump, so I should have it by 0800 tomorrow. I will have enough time (hopefully) to get it installed. If I am lucky I may even be able to do a fuel flow test. But as Murphy’s Law would have it, I was assigned a redeye trip for tomorrow night, so I have to go to work. Sounds like we won’t be starting the engine tomorrow.

FullSizeRender 9

Ignitions, Wire Routing, etc

The blog post is going to be quick tonight so I can get to bed. Today I started by verifying the spark plug connections on the ignitions. While doing that I decided that I wanted to swap one of the shorter and longer wires to allow for better routing. That turned into completely re-running all of the spark plug wires to try to give them a clean routing and not have any wires that end up too short or to long. I came up with a routing I was happy with.

Then it took a significant amount of time to secure all of those wires, fabricate stand offs where they route close to hoses, etc. In the end, I am pretty happy with how it turned out.

Once that was done, I went through the timing procedure for the P-Mags. To borrow a line, it is so easy, even a caveman can do it. The LEDs on the ignitions tell the whole story. I should note, that after my dissertation of my confusion over timing in my post yesterday, James left a comment on the post. Turns out that there are marks on the engine side of the starter ring gear as well, which are the ones you align with the case split. So either timing method will work on my engine, and I don’t need the special tool. So I timed the engine using that reference. Thanks James.


After I had the ignitions timed, I did the pull through test detailed in the P-Mag manual. Basically you verify that the proper plugs are sparking in the proper pairs and at the proper time. This verifies many parts of the system as well as the timing and installation. When I did this, the left ignition worked, and the right ignition did not work. (I am using aircraft left and aircraft right for a naming convention so it is opposite of a normal tractor airplane engine configuration.) I thought about it and did some troubleshooting. I verified the switch wiring by using a continuity check with each switch terminal. I verified all the wiring at the ignition. I also verified that the plug housings were grounded to the engine. I got to thinking what could cause one ignition to be completely dead, but store the timing, and show it was at top dead center the whole time. I had a theory, and called the manufacturer who knew exactly what it was. Apparently they have seen this problem before.

Turns out that I didn’t realize that my engine accessory case had a spacer for an impulse magneto. This spacer won’t allow the gear on the ignition to mesh with the drive gear in the accessory housing. So the ignition wasn’t spinning when the prop rotated. I had figured out what was happening via my trouble shooting (ignition isn’t spinning), but thankfully Brad knew about the spacer issue and could tell me exactly what I did wrong. The ‘educational” part of experimental airplanes is really the biggest piece.

So i removed the ignition so I could remove the spacer. Since there was a spacer, the mounting studs were extra long and had to be replaced so that the didn’t conflict with the ignition housing. All of this should be simple, but working in an 8 inch area between the engine and firewall around tons of wiring, hoses, and an engine mount makes it time consuming and uncomfortable. Something about only being able to get 1/8 of a turn on a wrench that is frustrating. I don’t know how this happened, but I had the exact length stud that I needed as well as a gasket. Thank god for keeping removed parts and ordering extra gaskets.

Long story short… ignition removed, remove spacer and gaskets, remove studs, install new studs, new gasket, mount ignition, re-seat ignition wiring connector and plug wires. Done.

IMG_3772 IMG_3773

Then it was re-time the ignition and do the pull through test again. This time both ignitions worked. All the plugs are firing in the proper sequence. So I am pretty convinced that all is well with the ignition system. Groovy.

Lastly we removed the fuel filter and disassembled it. It was pretty much clean as can be. I did find some debris in there, but it appeared to be a little bit of dirt or dust. It was in the threads of the filter housing, so I am not sure if it was present before fuel went into the filter or if it came from the tanks. But in any case, the amount is so minuscule as to not be considered at all. I will continue to check the filter often in the first hours of operation. But I think our efforts at keeping debris/dirt/dust out of the fuel system were successful.

IMG_3774 IMG_3775 IMG_3776

That filter photo is without any cleaning. The photo of my finger is the little bit of debris taken from the filter housing.

Oh, Almost forgot. We also tensioned the alternator drive belt. I just need to safety wire the bracket bolts.

For those of you keeping score at home… we have fuel, air, spark, lubrication…. whats left?

Slow Day, education, and more Fluid

Today started out very slow. I spent time learning about timing the ignition to the engine. The Lycoming manuals are not very descriptive because they are intended to cover a lot of different engines and accessories. To add to that, my combination of engine, ignition, injection, propeller, starter, etc, etc, etc is not something for which Lycoming writes a manual.

My biggest problem was I couldn’t quite figure out why Lycoming had two different timing methodologies. I didn’t know whether you could pick which one you wanted to use, or if the engine required a certain one. The difference is one method uses the case split to align the timing mark, and the other uses a drilled hole in the starter. Well, I have changed my starter and it doesn’t have a timing hole, so I was hoping that I could use the case split method. But I knew that being able to choose which one didn’t sound right, and it wasn’t. So I needed to figure out which method my engine requires.

My buddy Jeff Lange said, “Well, why not pull plugs and see if the cylinders reach top dead center with the timing mark near the starter or near the case split.” DUH. Seems simple. So I did, and I certainly need to time based on the missing hole on the starter. Turns out there is a way to do that with a piston stop and a angle wheel. It is actually pretty simple, and seems to me that it would be more accurate. I was able to borrow that tool set up, so maybe I will get to that tomorrow. (Hopefully I will remember to take some photos of it.)

By the time I learned about this most of the morning was gone. I did manage to make up a line for the mechanical fuel pump pressure relief and find an exit for it out the bottom of the cowl flange. I think I will flox the tube into the flange eventually to keep it vertical, but it is functional the way it is.


Then Greg arrived and we spent pretty much the whole afternoon securing plumbing and wiring. We put a wrench on every end of every hose as well as every fitting in the oil and fuel systems to be sure they were torqued down. We also made sure that hoses and wires were clamped to prevent unwanted vibrations, chafing, rubbing, etc while still allowing the relative motion between the vibrating engine and “fixed” points on the fuselage and engine mount structure. Part of this was using adel clamps, which can be a pain in the butt. We also made quite a few stand offs with zip ties and short pieces of hose. We still have a few more things to verify routing on and secure, but we are probably about 75% there when it comes to securing things in the engine compartment.


Note that in the above picture, we do need to add firesleeve over the Red Cube fuel flow meter, but it is not installed yet so that we can verify that the hoses and fittings are not leaking before we cover it up.

Lastly, Kevin and I spent a few hours putting fuel in the strakes for the first time. We started by putting 5 gallons in each tank, sloshing it around and draining it via the quick drain openings. This was intended to wash out the tank and bring any debris that may be in there out through the drain. We were pretty careful when we closed the tanks, and they have been closed ever since. I was pretty certain we wouldn’t get much of anything out of them, and we didn’t. In fact the only thing that ended up in our funnel filters were a few specks that appeared to be bugs that likely made the fatal mistake of landing in the funnel. Once it was drained, we installed the fuel quick drains in place of the plugs that used to be there.


After that, we put the fuel back in the tanks (again through a filter) and turned the fuel selector on. This brought fuel through the valve, filter, and electronic pump, through the fuselage plumbing all the way back to the firewall. We checked for leaks at any of the fittings in the system. We didn’t find any. 🙂

One of the leading causes of accidents on new homebuilts is a fuel delivery problem. As such, the FAA is getting pretty serous about having new homebuilders do a fuel flow test prior to first flight. So we did one. Basically the test has you put near minimum fuel in the tanks (we used 5 Gallons), put the airplane in a 5 degree nose up attitude, and check what the fuel flow rate is through the fuel system. The requirement for a successful fuel flow test for fuel injected engines is 125% of the maximum fuel flow the engine could require.

There is a formula and charts to figure out what this number is. But for my engine it was 2.08 lbs of fuel per minute. So we got the scale out and timed a minutes worth of fuel flow in the 5 degree nose up attitude with the fuel boost pump on. For the left tank we got 3 lbs 9 oz of fuel in one minute, and for the right tank we got 3 lbs 10 oz of fuel in one minute. That means my fuel system is capable of delivering 219% and 223% of the required fuel flow respectively. I would say that is passing the test with flying colors.

We put the fuel back in the tank. Tomorrow, I will check the inline filter to see if we got any debris that flowed through the system. If there is any signs of debris, we will flow some more fuel until the filter checks clean.

Air, spark, and starter 

Greg and I are continuing working on engine stuff. There is a relatively long task list. But we actually checked quite a few things off today. 

Greg spent a lot of time working on the air induction system. He got the alternate air door/filter mount fit to the firewall and drilled to match the mounting. That was no small job actually. There is still one more thing needs to happen to the alternate air doors. That is to install magnets to keep the doors closed in normal operation. It’s just a simple Flox job to add those. And I need to find the right magnets… Strong enough to hold the doors closed in normal operation, but weak enough to allow the engine to suck them open if the air filter gets clogged. 

In addition to getting the alternate air/filter mount installed, Greg got the flange mounted to the fuel injection servo and bridged the gap with a piece of scat tube. 

We still need to safety wire the duct flange on the fuel injection servo, but we will get that tomorrow. 

While he was working on that, I had to go back and change the gaskets on the ignitions. Long story, but I got them installed and the ignitions torqued down to the proper specs. 

Then I switched gears to do a bit of wiring. We are working towards a first engine start, and in order to do that, we need the starter operable. Eric is on a much needed vacation, so I had to see if I can put what I learned from him to use. I started by adding leads to the starter push button and then terminating these leads to a connector. So I soldiered on a pair of wires to the switch, added some heat shrink and labels, and then crimped on pins and assembled a connector. Incidentally, a spring clamp works well as a third hand. 

I didn’t take any pictures of the rest, but it was basically taking the starter lead from the VPX and terminating it in the mating connector to the starter switch, then running a wire from the other terminal back to the starter solenoid. I also needed to add a ground to the solenoid, which was a small jumper from the firewall ground bus to the base of the solenoid. Easy peasy. 

We then had a starter push button. We don’t have the starter ring gear installed, so I did go ahead and try it. Low and behold, the starter motor spun! Good deal. 

Greg and I then worked a little bit on the fuel vent system. There will be more on that tomorrow. 

We also worked on the bleed air system, which gives the pressure reference to the fuel injectors. Basically it is just a pressure reference port in the induction duct. In order to complete the system, we floxed a tube into the duct as near the fuel injection servo as possible. We then installed everything to make sure the tube gets bonded in the proper position. 


I also spent some time cleaning up some wiring and tying things off. Being sure to allow enough slack to prevent vibrational failure, but keeping things from chafing, or getting caught on things. In some places I used silicone fusion tape to further prevent any chafing.  There are plenty of more wires and hoses to secure, so there will be more photos to come. 

Lastly, I swapped two wires on the rocker switches to make the labels correspond to their functions. Last time they were installed, the master and the alternator switches were reversed. All better now. 

Tak about a good day in the Garaggio!