Online Community of Zenith Builders and Flyers
I started looking for an engine cowl for my setup over a year ago and couldn't come up with a solution, so I am now in the process of designing and building my own engine cowl. (Zenith 701/ULPower 350iS) I know this is a bit of an odd combination, but it should make for a Hot Rod Sky Jeep... :)
I almost have the male plug ready for the next step (see attached picture), but I am a newbie and I have questions:
*Is it worth the time and effort to build a female mold from this male plug?
*Vacuum bagging a good option?
*Should I just do a one-up production and layup the cowl material on the male plug?
*If I use the male plug for the final product, how do I clean up the layer overlaps and smooth the surface for paint?
*Fiberglass or Carbon Fiber?
*How many layers, what type, and what weight of fabric?
*Do you all think there would be any interest for building a few of these to sell?
This is just my concept and style. I'm sure it will take some fine tuning to get everything right, especially with the rear-mount oil cooler.
I have taken the EAA Composites class, but this is still a big project for a first timer. I made a post in the 701 specific discussion back before I started. There was some great information and I hope those guys will continue to share their knowledge here for everyone to see.
Also, here is a short video link to the spray foam process I used for the plug:
Thanks for looking and I really appreciate any and all feedback, suggestions, criticisms, or any other input you are willing to share!
1) In a factory making production parts, you would make a female tool to keep the OML smooth. But for a one-off, I'd do the layup on male plug because part will come off easier if you don't have lock-on where shape keeps part attached. Be sure your mold does not have lock-on. Also, you would like to have some draft on surfaces having indentations so part will release easier.
2) Vacuum bagging is better than contact layup if you know what you are doing. Bagging has tricks in making bag dip into recesses without bridging, If you have bridging, the radiuses will suffer from lack of flow from low compaction and have rough finish and places of weakness. I would not vacuum bag on first attempt.
3) I'd layup on the male mold. The difference in exterior mold will be where you have overlaps. For a simple cowl with low stresses, you can even butt plies and still have sufficient strength. You always stagger overlaps or butts by 1-3". Butts will avoid bumps and make exterior smoother.
When part is cured, you will be sanding and filling surface imperfections so it will be smooth no matter whether plies are overlapped or butted.
4)Sand cured surfaces and fill with Bondo polyester filler. Be prepared for multiple applications as each try will show mismatches and bumps.
5)Definitely 7781 fiberglass wove fabric, 10 oz wt. Glass is drapeable, cheap and has sufficient strength and stiffness for a cowl. I assume you are making layup monocoque, approx 1/8" thick. A sandwich would be lighter but more complex to build as first try. Carbon is stiffer, more expensive by 3-4X and abrasive when cured which will dull cutting tools unless they are carbide or diamond.
I was just about to post my reply and saw Bob's! I definitely agree - you yourself said the engine/airframe combo was an "odd" one, so since you're not looking at a large market and many cowls, I'd go the one-off 'glass over male plug. You're already mostly there since you have your male plug already. I also agree that it would be tough to learn the vacuum bagging process on something this complex!
As Bob points out, you have to fill and sand the cured glass, anyway, so the overlaps are no big deal and will smooth-out during the filling and sanding process. I've only made small parts such as elevator tips, gauge housings (for an automobile), and a small fairing for a wingtip ADS-B out device, and did all of these with Rutan bi-directional cloth (lays easily over compound curves) and West System 105 resin with the 206 "slow" hardener (gives a longer pot life/working time). I added glass micro-beads to the resin to make a lightweight filler for the filling and sanding process. However, sounds like Bob has been there/done that with a cowl so I defer to him on materials and methods!
When it comes to surface fillers on glass, you can use laminating epoxy resin with glass bead filler, as John suggests, but the epoxy makes a harder finish than Bondo polyester and is more difficult to sand. The epoxy based filler would have a slightly less tendency to crack after 15-20 years but by that time, you would be looking at refinishing anyway. I've done both fillers and prefer using Rage, a Bondo-like polyester.
When using epoxy resins, you want to be aware of exothermic reactions caused by mixing too large a batch of resin at one time. About 1 cup is a good starting point and using a slower reacting hardener is a good idea because that lowers reaction energy produced. Another trick is to have the 2-part materials in separate cups unmixed, then mix them just before needed. This keeps the mass of active resin low and avoids rapid exotherming.
You want to be sure that the mold surface is well released so the laminate does not stick to it and will come off after cure. PVA is a material that coats the mold and is water soluable and releases easily. You can use the old Simonize car polish made with carnuba wax as that also releases.
The expertise available here is amazing! Here are my thoughts at the moment:
I still have a couple more light sanding/fillings with the foam plug's "hard skin" to finish removing surface imperfections and get it just the way I want it.
I will paint it with some left-over automotive paint for a glossy, smooth, and clean finish. Then I will wax it several times and might also spray it with PVA, just for the extra assurance that it will release.
I will probably not use the foam core construction unless it is significantly better for a cowl. I could still be easily persuaded to go either way here. If a foam core is used, how do you smoothly merge it into the cabin sides without a thick "step" at the firewall? Also, how do you make the tighter contours, such as around the intake openings?
I plan to prep and lay-up with ~10oz fiberglass and West System slow cure epoxy.
How many layers will it take to get a thickness of about 1/8" for the cowl?
About how much fabric and resin do you all think I will need to complete this job?
Do you recommend a peel-ply on either side like we used in the EAA composite construction class?
I will probably use the epoxy and micro balloon as the filler for fixing the surface imperfections once the cowl is complete. This airplane will very rarely be outside when it isn't being used, so I would hope the paint will last as long as I am able to fly. We have had good quality urethane on vehicles for several decades that were garage-kept and they can be cleaned up and still look amazing.
If the plug is still in good enough shape, I might go ahead and use it one more time for a second cowling before I destroy the mold during removal. Is this practical or even possible? If everything goes and works exceptionally well, I would have a second cowl as a spare, to sell, or even as a possibility to create a female mold from later on.
I know there are a lot of questions here.
Thanks again for helping a composite newbie and please continue to give me any thoughts and suggestions you all may have!
Don't forget that your male mold makes the IML (inner mold line) of the cowl and will not be seen so a super finish may be overkill.
- The auto paint will be good to fill pores and your release coatings should work.
-Sandwich construction is intended for weight reduction as it provides a stiff surface with minimum weight. It requires multiple steps in construction. I would use it for my cowl, however, a conventional monocoque wet layup is faster and easier.
-Each ply of 7781 is .010 " thick so for a 1/8" thickness, you need about 10 plies as the resin will increase the final per ply thickness. The glass weave for best drape is 8 Harness Satin which comes in 50" width, AC Spruce P/N 7781-50, $9/yd. Just guessing at the quantity for your cowl, I'd say 10 yds of glass.
-You don't need to have a continuous wrap around cowl. You might divide the wrap into 3 or 4 sections to make draping easier. Then the next wrap will stagger say 2" from first so the joints don't line up. You would keep applying ply sections until you build up up the total thickness of 1/8".
-Peel ply is used for several reasons: to absorb excess resin, to maintain a clean surface for secondary bonding, to provide a textured surface. I would use it on the exterior surface only.
The mold you made looks great!
The guys that build the Zenith 701 cowls for the smaller ULPower engine came out to my place about 8 months ago and brought a 260 engine cowl that they could cut up, measure, mark, mock-up, and modify to work with the bigger 350 engine. I still have pieces of that cowl. It was all fiberglass and was only about .060" thick. I recently asked what they are using on another plane cowl they are working on and here is the response I got:
"As for making the part it is 2 layers of uni infused between 2 layers of bid with a core material in various locations as needed. Every cowling is different and the layup is based on shape, size and aircraft type. On this cowling it will be carbon to do it in glass we would have to do an additional layer of glass."
What are your thoughts on this? Is 10 layers a bit overkill for the 701 Sky Jeep. Would maybe 8 layers be adequate or do I need to stick to the 1/8" layup schedule?
Also, should I use UNI and BID fabric, or just BID rotated 45 degrees for each layer?
but it should make for a Hot Rod Sky Jeep... :)
Totally off topic, I know. Also, not helpful, I know. But I just gotta ask - since the 701 airframe has Vne of 110 mph, is "hot rod" an apt description? Maybe rocket - since it will climb like a shuttle launch!
Hmm, good question. I've always considered "hot rod" eligibility based on how fast it could get to 100mph. If this is the case, this combination should handle the description quite well. I could be entirely wrong with my thought process though.
I have no data to back this up yet either, but I am getting into the final stages of construction so hopefully it won't be too much longer.
Always make a negative mold of your male mold. You put a lot of energy (and money) into your mold. If anything goes wrong with your cowling or you want to make modifications without working on the original then a new cowling is quickly made.
Or you could produce a cowling for someone who could use a similar one. you might be interested in the following links:
The duratec will give your plug and also the negative mold a perfect finish. This product can be easily sanded and polished.
For the negative you can use fibreglass mat and polyester instead of the more expensive glass woven fabric and epoxy wich you will use for the final product. Someone recommended to me the use of vinylester instead of epoxy because it is more heat resistant than epoxy.
Keep it simple and use lay-up method. Vacuumbagging is rather complicated and you need a vacuum pump and quite some experience plus a lot more vacuum materials which are wasted.
Hope this will be of some help.
By the way, very nice design your mold.
Vinylesters have lower Tg (softening temp) than epoxies so I would stay with epoxy as matrix resin because cowl can get warm and you don't want distortion.
The plies at 45 deg are intended to provide shear strength and plies at 0/90 deg are for bending resistance.
If you build the cowl on the thinner side without stiffeners and it flexes during flight, you may wish it was thicker.
From a composite engineering design, you would build a cowl with either sandwich or a skin/stringer design for lightest weight. As I said before, if you haven't worked with composites, keep the layup simple but with sufficient strength margin. A workable ply orientation would be 0/90,0/90, +-45,0/90,0/90 +- 45, 0/90,0/90. Wet layups tend to be resin rich so the per ply thickness will be around .013-.015" thick, bringing the laminate final thickness to ~ .10".
Just use the woven fabric and forget uni plies as you are not trying to optimize the design for weight.
Thanks for the compliment on the design. I have never seen one quite like it before and I just hope it will all come together and work like I think it should.
I am hoping to make a second cowl that could easily be turned into a mold later on if needed, but we will see how the first one goes and if I am still ambitious about having a second cowl at that point.