• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

Glow to electric conversion: 201 Level Course

willsonman

Builder Extraordinare
Mentor
#1
So I know I've covered gas/glow to electric conversions before on these pages but I've got a new one that I am going to let out of the bag. and cover it here for those who really want something a little more complicated. -- a twin engine aircraft.

Now, there's nothing really much of a mystery about this process other than the fact that you are doing everything twice. I also realize that there may be a few things that are unique to this particular project that may be insightful for some folks. Each post will have steps associated with it in the order that I perform them. The subject we are dealing with here is a Hangar 9 B-25 which has been out of production for 13 years and this particular model is 15 years old now. Previously flown on Evolution 36 engines the model came to me with those engines, a full suite of servos (9 I think), an old Futaba receiver I cannot seem to even give away, and a receiver battery that was about as dead as you can get. With all that context, let's get onto the process.

Step 1:
With the model having engines installed I opted to try and get them running again so that I could sell them to offset the cost of the project. $100 for both engines goes a long way here. I removed the props and cowls and soaked the carbs in nitro fuel overnight and they were still stuck but a hit with automotive carb and throttle body cleaner freed one while the other required a bit more persuasion. Eventually free, I reinstalled the props and put fuel in each tank and primed the lines per usual procedures and they both fired right up. What was surprising was that they both were well-synced up too.

Step 2:
I found a buyer pretty quickly for the engines which meant I could proceed with removing the engines. This is a simple process of removing one screw or bolt at a time. I removed the fuel tanks as well as the throttle servos. The tank removal required that I break away a piece of balsa holding them in position as part of the original installation. For this model I did have to remove the nacelles from the wing but that's two simple bolts.

Step 3:
Clean. Then, clean something. Then, clean some more. Even though this model was not put away "wet" per se, the model still had a film of oil from the glow fuel and then sat for a long long time and accumulated dust on top of that oily base. Combine this with the stink of cigarette smoke and you have yourself quite the cleaning project. Along the edge of the nacelle here you can see the line between the clean nacelle and the dirty wing. Then look at the solvent spray onto the wing and how the solvent is just melting away the grime. I'm using paper towels and Simple Green but really any good degreaser will work here. Kitchen degreasers like 409 work well too. Even after the initial cleaning I went back and did it all a second time with the degreaser again, and then a final wipe down with a cotton cloth and windex. Then I let the whole thing air out and bake in the sun to help remove the smoky smell. The more sunlight you get the more this will go away. Short of putting the model in a bag with an ozone machine, there's really no good way to remove smoke smell. Fly it more and the smell will eventually go away.
IMG_1194.jpeg

Step 4:
The blind nuts in the firewalls were removed by screwing a bolt back into them and tapping the bolt with a hammer until free. The remaining holes were plugged with oak dowel from the hardware store. This model uses a ply back plate behind a fiberglass nacelle but this same technique applies universally. I did have to drill out the holes to accommodate the larger diameter of the dowels but you kind of want that for a nice and snug fit. I then tapped each dowel flush with the firewall and soaked each one with thin CA until it pooled on the surface. It's a lot so I recommend a fresh bottle if yours is low. Some folks swear by epoxy for this job but CA has always worked well for me. The idea here is to plug the holes to maintain structure but also provide material that you may need to drill into again for new holes for your motor mount.

I'll note that I did this step along with the cleaning so note the cleaning marks for entertainment purposes for how dirty this model was.
IMG_1196.jpeg IMG_1198.jpeg

Step 5:
The last part I'll post here is sourcing replacement parts. Sometimes you are lucky and sometimes you are not. In my case, I'm both. New cowls are just not available. The ones I have can fly the model fine but they have a large hole for the head and muffler to protrude as well as holes for the adjustment needle and fuel filler. The Front greenhouse plastic is pretty beat up and the other plastics are not in the greatest of shape but are fairly serviceable. A quick hop over to Park Flyer Plastics gives me a COMPLETE set of BRAND NEW vacuum formed plastic parts for $17. Shut up, and take my money. New parts all around will keep the look consistent rather than replacing just one part and the UV damage over time is a real thing on old models.
 

willsonman

Builder Extraordinare
Mentor
#3
Interested to see how this goes!
I've got a pretty good method down and I've been flying my first conversion for a while now. ;) 6 years of flying the snot out of that airplane. While my current iteration of that airplane has a more scale 3-blade prop the approach is the same. More on power system selection later, which can be a highly debated topic. Again, this is my approach, not THE approach.

edit: Correction, the Hawk was my second conversion. I forgot about the Top Flite Texan I did and then sold to @earthsciteach. I may have missed another that escapes my memory.
 

cyclone3350

Master member
#4
[QUOTE="willsonman, post: 719902, member: 457"


Park Flyer Plastics gives me a COMPLETE set of BRAND NEW vacuum formed plastic parts for $17. Shut up, and take my money. New parts all around will keep the look consistent rather than replacing just one part and the UV damage over time is a real thing on old models.[/QUOTE]

I've been ordering some parts here & there from PFP the past couple of months. Their prices R so reasonable, that I'd order several of different sizes just to see what works & still do alright in total $$. I get the parts in a week or less & have no issues with their quality. Banana Hobby has a B-25 that is very close in size as the H9. Might be a source of plastic parts there. WS 79" LT 62.4". Did an order from them not too long ago & they were pretty quick also.
 

Tench745

Elite member
#5
So I know I've covered gas/glow to electric conversions before on these pages but I've got a new one that I am going to let out of the bag. and cover it here for those who really want something a little more complicated. -- a twin engine aircraft.

Now, there's nothing really much of a mystery about this process other than the fact that you are doing everything twice. I also realize that there may be a few things that are unique to this particular project that may be insightful for some folks. Each post will have steps associated with it in the order that I perform them. The subject we are dealing with here is a Hangar 9 B-25 which has been out of production for 13 years and this particular model is 15 years old now. Previously flown on Evolution 36 engines the model came to me with those engines, a full suite of servos (9 I think), an old Futaba receiver I cannot seem to even give away, and a receiver battery that was about as dead as you can get. With all that context, let's get onto the process.

Step 1:
With the model having engines installed I opted to try and get them running again so that I could sell them to offset the cost of the project. $100 for both engines goes a long way here. I removed the props and cowls and soaked the carbs in nitro fuel overnight and they were still stuck but a hit with automotive carb and throttle body cleaner freed one while the other required a bit more persuasion. Eventually free, I reinstalled the props and put fuel in each tank and primed the lines per usual procedures and they both fired right up. What was surprising was that they both were well-synced up too.

Step 2:
I found a buyer pretty quickly for the engines which meant I could proceed with removing the engines. This is a simple process of removing one screw or bolt at a time. I removed the fuel tanks as well as the throttle servos. The tank removal required that I break away a piece of balsa holding them in position as part of the original installation. For this model I did have to remove the nacelles from the wing but that's two simple bolts.

Step 3:
Clean. Then, clean something. Then, clean some more. Even though this model was not put away "wet" per se, the model still had a film of oil from the glow fuel and then sat for a long long time and accumulated dust on top of that oily base. Combine this with the stink of cigarette smoke and you have yourself quite the cleaning project. Along the edge of the nacelle here you can see the line between the clean nacelle and the dirty wing. Then look at the solvent spray onto the wing and how the solvent is just melting away the grime. I'm using paper towels and Simple Green but really any good degreaser will work here. Kitchen degreasers like 409 work well too. Even after the initial cleaning I went back and did it all a second time with the degreaser again, and then a final wipe down with a cotton cloth and windex. Then I let the whole thing air out and bake in the sun to help remove the smoky smell. The more sunlight you get the more this will go away. Short of putting the model in a bag with an ozone machine, there's really no good way to remove smoke smell. Fly it more and the smell will eventually go away.
View attachment 225213

Step 4:
The blind nuts in the firewalls were removed by screwing a bolt back into them and tapping the bolt with a hammer until free. The remaining holes were plugged with oak dowel from the hardware store. This model uses a ply back plate behind a fiberglass nacelle but this same technique applies universally. I did have to drill out the holes to accommodate the larger diameter of the dowels but you kind of want that for a nice and snug fit. I then tapped each dowel flush with the firewall and soaked each one with thin CA until it pooled on the surface. It's a lot so I recommend a fresh bottle if yours is low. Some folks swear by epoxy for this job but CA has always worked well for me. The idea here is to plug the holes to maintain structure but also provide material that you may need to drill into again for new holes for your motor mount.

I'll note that I did this step along with the cleaning so note the cleaning marks for entertainment purposes for how dirty this model was.
View attachment 225214 View attachment 225215

Step 5:
The last part I'll post here is sourcing replacement parts. Sometimes you are lucky and sometimes you are not. In my case, I'm both. New cowls are just not available. The ones I have can fly the model fine but they have a large hole for the head and muffler to protrude as well as holes for the adjustment needle and fuel filler. The Front greenhouse plastic is pretty beat up and the other plastics are not in the greatest of shape but are fairly serviceable. A quick hop over to Park Flyer Plastics gives me a COMPLETE set of BRAND NEW vacuum formed plastic parts for $17. Shut up, and take my money. New parts all around will keep the look consistent rather than replacing just one part and the UV damage over time is a real thing on old models.
Consider me enrolled; I like learning.

Park Flyer Plastics is awesome. I ordered some cowlings from them for my Dumod build. Sturdy plastics, well formed, and affordable. Plus, Keith Sparks is an avid foam builder (which you know).
 

willsonman

Builder Extraordinare
Mentor
#6
Consider me enrolled; I like learning.

Park Flyer Plastics is awesome. I ordered some cowlings from them for my Dumod build. Sturdy plastics, well formed, and affordable. Plus, Keith Sparks is an avid foam builder (which you know).
Yeah, he's been around the block and has literally written the book on foam model construction.
 

willsonman

Builder Extraordinare
Mentor
#7
Moving into our next phase of the conversion is the idea that we have to eventually put motors back into the airframe. This particular model had provisions for electrics but not in a usual way. It used aluminum standoffs and nothing really notes how long they were and they were to be used with a very specific motor. No use in trying to go that way if we can be a little more flexible.

Step 1:
Determine the depth of your motor. This is defined as the distance from your firewall to the front plate of the prop adapter, or the back of your propeller. This allows proper distance to clear the cowl when everything is reinstalled. Generally you can find this in the product manual or plans if it is kit built since the glow motor has the same requirement. I prefer mm for ease of accuracy. With this distance we can now start the process of motor selection.

Step 2:
Selecting a power system is a pretty hot topic. I'm a firm believer on lower kV, Lower cell count, and larger prop diameter. With this model we are fairly constrained on the prop size since we are fairly low for ground clearance as well as side fuselage clearance. I opted to just focus on the stock recommended power system sizing with my own principles applied. I am targeting around 125W/pound so the recommended AUW of 15.5 pounds means about 2000 watts is required with 1000W on each motor. The recommended Eflite Power 46 falls just shy of this at 800W and is tremendously overpriced IMO. So, let's take that 670kV specification and see if we can find something a little better for power and price.

Enter the SunnySky X V3 series from BuddyRC. I have been tremendously impressed with these motors for their price, efficiency, smooth operation, and reliability. They also pack some serious power capability. The X3120 V3 was selected with the 760kv option. This provides the 1000W target with a bit higher kV and should support more scale 3-blade props at 13" diameter. Couple this with the ZTW Gecko 65A ESC and we have some seriously good parts to install. Cost: my selection is $208 compared to the recommended setup with the eflite motors alone costing $220and then another $196 for the recommended ESCs. So, we are literally HALF the cost with more power, higher RPM, with high quality gear. That's a win in my book.

Step 3:
With the motor selected and the depth requirement we can assess how we will mount the motor. Sometimes you just need a spacer or even a couple of washers. IN our case, we needed about 2 or so inches. My default mounts for this situation are the tried and true brushless motor mounts over at Horizon Hobby. Formerly Great Planes brand (now retired) they are branded as Spektrum parts and they are drilled to meet up with their motors. Expect to do some drilling here to make the mount meet up with your motor. A drill press helps but is not required. More on this later.

Step 4:
Order parts and wait.
 

willsonman

Builder Extraordinare
Mentor
#8
If you are curious about some more specifics on my approach to motor and prop selection, feel free to watch this video on my thoughts:
 

willsonman

Builder Extraordinare
Mentor
#9
For our next lesson we are trying to mount the motors and start to get a rough idea where the batteries are going to land. Yes, the manual for this airplane says to put the batteries under the belly pan in the fore most position. I've learned to trust these recommendations, but I would never fly without verifying them. Let's proceed.

Step 1:
The downloadable files include not only the manual but also firewall templates, which provides provisions for marking for glow or electric power. Why the difference between the two? It has to do with symmetry. Glow motors generally do not mount on a perfect square, unlike the X-mounts for our electric motors that we are used to. Then factor the rotation of the glow engine to have the cylinder in a particular orientation to access the glow plug and mixture needle. This also correlates to where the throttle servo is intended and connecting the push rod in a sensible way.

So yeah, print out the template, cut it out and hold it up to your firewall. In my case, I held it up and it looked off. Turns out there was a little bit of scaling to fit the page better so I made SURE there was none for the next round and that cutout fit much better. I like to use sharpie on these smooth surfaces because it works well to mark them but then you can use a little bit of rubbing alcohol on a paper towel to remove marks later or remove them because you marked them wrong. Then you can use the template provided with the mounts to center on your cross marks and drill for blind nuts. These are the SAME blind nuts I removed from the glow motors. Waste not.
IMG_1339.jpeg

Step 2:
Mount the mount. So the mounts come in two halves and this is honestly the easy part. Just use the bolts you've got from the glow mounts and bolt it to the firewall. Orientation really doesn't matter here. If the bolts are bigger than your mount holes you can drill them out but be careful to keep them center. You may want to try using a step bit for this if you do not have a drill press. I do not recommend just drilling it out with a hand drill and a bit. The hole WILL be off, I don't care how superman you think you are. Of course the other recesses that are being used here are not holes so I recommend a carbide bit on a dremel tool to open that up a touch if you need to. As-is, mine fit just fine.
IMG_1340.jpeg

Step 3:
Here's where things get interesting. The provisions for the front half of the mount have holes for a variety of motors but if I'm honest, they have NEVER fit ANY of the motors I've used with them and I have always had to drill my own holes. I was not much more lucky here but I did have a 50% success rate as two of the 4 holes did match with pre-drilled holes. SO I had to drill additional holes that basically were right next to other pre-drilled ones, and so I just wallowed out the fraction of material between them for an elongated hole. Now, the next problem is that you've usually got screws that are designed to be counter-sunk. You have two options. 1) get new screws or 2) countersink your holes. Just a word of caution here: If you get new screws just make sure they are not too long or they will go into the windings of your stator and cause a short, ruining your motor. I have a countersinking bit so I just countersunk my holes and installed the motor.
From here, I referred back to that mounting distance. Remember that? Yeah, we still need it. My rulers all have a little bit extra at each end that is not graduated so I cut a piece of scrap oak dowel (the same stuff used to plug the firewall holes) to the length I needed for that depth, which was 4-3/16" or 107mm. Using that dowel I was able to install the motor mount halves and set the depth of the motor.
IMG_1342.jpeg

Step 4:
Assembling the wing to the fuselage is a 2-bolt affair but this sucker is a bit big and awkward in the shop. I removed the nose greenhouse and battery compartment lid. There was 8 oz of lead stuck in there that I removed and that was my first clue that the stock location of the batteries did not seem right. Using a variety of battery sizes I played around with the stock location but it ALWAYS came out tail-heavy. OK, So we are not going with the stock location because that would require ballast I'm not willing to put on. The CG was a touch tail-heavy still with batteries in the cockpit area but a big 4S 6600mAh pack right in the nose gave me a DEAD ON CG. Some veterans of FF may recall these packs. Mine is still kicking and it IS a big pack... so yes, it IS a big airplane. So, at roughly 3300mAh per motor here I'm looking at my stash for how I'd prefer to power this airplane. While a single BIG pack may be more reliable, smaller packs are more economical and may prove more flexible for installation. Remember, just because the batteries you have on-hand now fit does not mean that those same packs will be available to purchase again when they need to be replaced. Think long-term here. I prefer to fly a bit more nose-heavy anyway so I'll stick with more round-numbered 4S 4000mAh packs.
IMG_1346.jpeg

With this battery location decided I need to extend wires on ESCs... more hotly debated stuff there too. So wire has been ordered because I do not have enough on-hand for the job.

Questions and comments are welcome.
 

willsonman

Builder Extraordinare
Mentor
#10
I should note that I did put the props, cowls, and ESCs on "loosely" (LOL) just to get a more realistic look at the CG. You can see that in the picture but I neglected to mention that.
 

willsonman

Builder Extraordinare
Mentor
#11
Been a bit occupied on a number of fronts but I want to mention that my wire and butt joint solder tubes arrived. I have what I need to make the ESC extensions now. Hard to say when I'll get to more updates here but I really want to push forward on it when I can get some shop time.
 
#12
Got a glow plane that doesn't have a motor in it. Goldberg eagle 2. I have a 5055 motor that would be way too overpowered for it, but would be cool. The motor is so big it takes up nearly the whole area where the glow engine would go. Would be cool though.
 

willsonman

Builder Extraordinare
Mentor
#13
I had an Eagle II years ago. Amazing airplane that does all the fun things. I flew mine on a 4020 670kv motor on 4S. If I nursed it I could get around 15 minutes of flying time on a 5000mAh pack. It was STUFFED in the nose but it worked.
 

willsonman

Builder Extraordinare
Mentor
#15
I suspect you'll discuss it when you do your ESC lead extension, but I'm curious about the practice of adding capacitors when extending power leads to ESCs.
I will but I'm fine to mention my thoughts briefly now. For the most part, it is my experience and opinion that you generally do not need to add capacitors unless you are in a very high-current application... say >150A. The entire point is to also only add capacitors that are low ESD (electrostatic discharge). This type of capacitor has a slightly higher discharge rate and eliminates some loss in our already fairly inefficient system. These capacitors are notably more expensive. Then there's where to put them. Always close to the ESC. Always. The idea is to reduce or eliminate the ripple in voltage caused by longer wires due to inductance in the longer length of wire. This ripple can cause the ESC to use the MOSFETs more frequently and cause them to "wear out" faster. In our lower-current applications there is plenty of voltage to supply the required current needed. Unless you are in a low voltage/high current application, there's not a whole lot of need. If you are in a high-voltage/ high current application there may still be need as voltage loss can become a bigger problem at higher voltages. These losses are why we use AC in our homes and not DC.

The short answer:
It's a very real problem but we are not talking about MILES of wire here. Feet is one thing and a few added inches is another. I'm looking at adding 18" of overall length of wire and I will not be adding capacitors.
 
#16
Some receivers like the Lemon Stabilizer ( if you use all channels ) receiver are recommended to install a capacitor, as the stabilizer may be using all channels at the same time for corrections and stressing the BEC into a brown out. So there is a time and place for a capacitor.
 

willsonman

Builder Extraordinare
Mentor
#17
Some receivers like the Lemon Stabilizer ( if you use all channels ) receiver are recommended to install a capacitor, as the stabilizer may be using all channels at the same time for corrections and stressing the BEC into a brown out. So there is a time and place for a capacitor.
I would argue that you likely would need to upgrade your BEC in that instance rather than add capacitors. Capacitors are only meant to reduce ripple in the voltage and are not in any way going to provide any additional current. If you are having brownouts in any application due to low current, you should upgrade your ESC with one that had a higher-rated BEC or change over to a dedicated BEC for your servos. The more complex a model gets, the more you would actually want this. The third option all together is to have a dedicated battery and just eliminate the BEC altogether.

All of these options have their advantages and disadvantages but they address the issue which is that the system is not supplying the needed current. Ohm's law wheel is your friend here. Capacitance is not a battery.
 
#18
I would argue that you likely would need to upgrade your BEC in that instance rather than add capacitors. Capacitors are only meant to reduce ripple in the voltage and are not in any way going to provide any additional current. If you are having brownouts in any application due to low current, you should upgrade your ESC with one that had a higher-rated BEC or change over to a dedicated BEC for your servos. The more complex a model gets, the more you would actually want this. The third option all together is to have a dedicated battery and just eliminate the BEC altogether.

All of these options have their advantages and disadvantages but they address the issue which is that the system is not supplying the needed current. Ohm's law wheel is your friend here. Capacitance is not a battery.
Years ago when I started to add the gyro or a gyro RX to my park flyers, I was having that issue. A club member noticed that my BEC's on my ESC where 2A & recommended what U just posted. Went with an Opto & added a 5a BEC from VH to give that a try. Bingo, it worked.
 

willsonman

Builder Extraordinare
Mentor
#19
While I continue to work on other higher-priority projects on the side, I thought I'd address a part of these conversions that typically arrives and it's generally not a pretty picture. The holes on the cowls that are made for protruding motors is one of my biggest complaints about IC engines in our hobby. It entirely ruins the look of a model when this happens. Even worse, it stops a conversion dead in its tracks because, well...

278445903_521725119657887_4214001891391570127_n.jpg


Yeah, it's not pretty. Needing to fill this void is a very real problem. Now I could go through and add fiberglass, blend it with bondo, and do a TON of work to get this cowl back in one whole piece... or I could buy a new cowl. Sounds like a no brainer right? Considering this model was discontinued 13 years ago, there are no cowls to be found. There just aren't.

So we need to dig into our creative toolbox and see what we can come up with.

There's been a massive push to start printing entire airplanes over the past several years and while it's gained some traction I do not ever see it becoming mainstream. Most guys I've talked to about it say it was a novelty thing they wanted to try but won't ever again. This is not an isolated conversation as I have had the same one with multiple folks. Where I do see traction being made is in aftermarket applications here. For example, the cowls. Now, I could have taken the time to draw up a cowl in CAD that would be really close in size and shape to the cowl I have... or I could spend $10 on a set of files from 3D Lab Print and try printing one that's already been done.
280136101_686522529274334_3876494926001045225_n.png
Now I did have to take this into Fusion 360 and get some measurements to figure out how much to scale up the part by. The file is meant to be thin-wall printed so it's what we call an "open" model, so I took it into meshmixer and closed it. We know we need nose weight anyway and the extent that we are scaling this up we will want the structure anyway to keep the print rigid. I'm doing this in ABS so the risk is very high that I'll have a print failure from some sort of temperature fluctuation warping the print but knowing I'll need to do some sanding anyway, I can use some spot putty to fix those imperfections.

This approach MAY NOT work but It's minimal cost, effort, and time on my part to NOT try it. ABS will hold up to the heat and will be resilient enough for the vibrations as well. It's worth the minimal effort to at least attempt to get at something somewhat presentable with work later to make it looking like it's NOT a printed part.