it's legacy firmware that runs on an Arduino zero but no s.port.

hello jim and all the crashlisteners , did you succeed in wiring and plugging the final touch from the xiao to the solenoid and ware you able to drive it by the mosfet?
i'm still stuck in the final since i haven't received the new "original vishay mosfets thinking the first batch received were ony fake or defective ones, or my wiring is wrong

hopefully we have 3 days in france and i have a few brutus full sets to my the breakin of my old reconditionned 4strokes and the plane will be completed this evening or so!
fxdr

Thanks for the explanation Dave. I did not even know there was an "Arduino Zero" before this!

Not yet, but I will do so in the next few days. Maybe today even.

You could log onto one of the dedicated electronics forums and ask how to test a MosFet in your preferred language. If you choose to do this, please provide a link here; some of us might be able to assist.
Pictures of your construction might help the reader to make a meaningful comment also.

Jim.

I wish I could post an attachment but I guess my post count is too low. I got the solenoid system working to the extent that I was able to start my engine (OS .15 LA) and run it and even do a really rough dial in of the fuel map.

immediately I have issues with my solenoid sleeve. It is resin printed on a cheap resin printer and I’m getting bubbles on the carb side of the housing. I guess it has to be machined or at least printed on a better printer like a form labs - or I need to walk in my tolerances more.

i am full stop until I can machine the sleeve but I may tinker with trying to print another one. I hope my solenoid isn’t somehow damaged or defective. I only get bubbles when the solenoid is running so I’m guessing air is coming in past the o-ring.

https://cimg9.ibsrv.net/gimg/www.rcu...4649d83d4.jpeg

It would be amazing if we could print housings like this. It would open this to people everywhere. Did you try to determine where the air leak is?

I did try a bit. I think it is air getting past the o-ring and mixing with the fuel in the upper chamber. The when the solenoid runs it pushes the air into the fuel line.

I may experiment with some bigger thicker o-rings. If there is some slop and leakage between the upper and lower chambers of the housing - it probably won’t make a big difference if the leak is really small relative to engine fuel consumption. The fuel map will be tuned to account for this. But if air gets past the o-ring, game over I think.

Maybe some RTV would help? The air could only get in at the tube connections or the top of the solenoid.

I can look at this too. I considered epoxy. Maybe some resin around the top and photo cure it. I might try that as a quick fix but I will need to de-grease the solenoid first.

Failure to update new receivers!
 

I'm not totally giving up......yet! I am now assembling a controller as described by Raleightcopter over on GitHub.
I am aware that I have limited opportunities to tune up while the aircraft is in the air etc.
Very good; I am somewhat relieved.
No need to apologize as I understand your motivation in the earlier post on this matter.

However, I apologise for the late reply as I have been thinking the subject through, at the back of my mind so to speak. I've not been sure how to answer until now.
Jim.

No Worries, Jim!

Only quoting the last post, but adressing all of the recent ones:

If needed, I can send you a housing, to see if that fixes the issue. I will be able to ship in about 10 days from now, and most likely cost would be around 20~25 Euro (22~27 US$ approx).
But it is way more interesting to see where you get with the 3D printed solution, so I'm not pushing :D

You could use sealant, but be aware that some sealants (silicon based usually) contain solvents that are acidic and MIGHT attack the metal parts in the solenoid.
Also, take care that no sealant ends up in the valve.
I agree on the leakage between the chambers, provided it is small in relation to the engine consumption, is relatively unimportant, as long as it is constant and does not deteriorate over time.
Leakage along the O-ring can be a problem, but in case muffler pressure is used, that should leak OUT, not in, since the tank would be under positive pressure, and only at low throttle settings (=> low tank pressure) this could be an issue.

However: at really low fuel consumptions, the valve tends to cavitate a bit, usually recognizable by the fact that visually the vapour bubbles have the appearance of a mist, or "dustlike"... This normally should not affect running of the engine too much.

Then keep in mind that your engine is bushed, AND really small, and not that long ago, another poster had issues with an OS 25LA. I believe those issues can stem from heat generation in the bushed front bearing, to which the carb is attached. This can cause vapour forming IN the carb, and coming backwards into the fuel line. This is a very confusing thing to observe sometimes.

I think the possibility to 3D print the housing is pretty neat, and if possible, I would like to ask to test this on a slightly larger ball-bearinged engine, to see if the issues are indeed coming from leakage, or that they are due to valve cavitating and/or front bearing heat generation.

Much appreciated input. Having slept on it, as much as my tots let me sleep, I am slightly MORE confused at my issues from yesterday. Muffler pressure should cause fuel to leak out of the upper chamber. Further confounding this is the fact that I was using crank case pressure with a 1-way valve to the tank and I was still seeing bubbles coming from the solenoid housing. I had a needle valve on a pressure relief vent on the fuel tank so I could tune back the tank pressure a bit as it was substantial. Hopefully my next run will have an integrated pressure sensor so I can log tank pressure but we shall see...

At any rate, the leak really doesn't present like it is coming from the solenoid seals but more likely somewhere else. I think I should do as you suggest... I have an MVVS .77 that I ran with no solenoid through about 200 mL of gasoline just to start the break-in process. I was able to control the mixture using an aquarium air pump hooked to an ESC (needle valve on the tank vent) and a knob on my transmitter. This engine was actually a lot easier to work with compared to the tiny .15 LA. I will revise my solenoid sleeve today and try again when I find time (who knows when that will be). I think I want to beef up the areas around the fuel nipples and actually go to larger fuel line so I can use a larger OD fuel nipple on the sleeve. I wonder if I am leaking in that area....

I just got notice that my desktop CNC router has shipped. It may not be able to get a perfect finish in aluminum for these sleeves (I didn't buy it for this) but I am sure I can find some thermoset polymer material that would work fine. Are you using a lathe?

I have no experience with crankcase pressure, but a great part of the successfull operation of the system is still in the mufflerpressure rising and falling with the RPM...
You really do NOT want a tank pressurized by a non return valve. Example: you run the engine at idle and adjust mixture. Transition to full throttle, tank pressure rises, you adjust mixture, OK.
Return to idle, and the tankpressure will NOT drop... Now your mixture is far too rich. If you adjust mixture, you cannot get the thing started next time because there is no tankpressure yet.

I strongly advise against crankcase pressure and non return valves. Muffler pressure is best.

Is your current carb made out of metal or plastic? IF you have a fitting Perry carb, try that. It's polymer, and does not heat-soak.

You will most definitely LOVE that size 2-stroke on the solenoid... Relatively very easy to set, and very responsive...

Here's my ASP .90 2-stroke on the solenoid system with a tuned pipe:

Mine are CNC produced by a company specialized in small series fine metalwork.

That ASP sounds SWEET!!! Does it have a keyed thrust washer? My MVVS has the collet-style thrust washer which is still usable for spark ignition, but every time I move the prop I will have to reverify timing... PITA.... I was thinking of just using the OS glow plug for gas... I suspect you get a better running engine on spark though but I lack experience to really say.

You are spot on about crank case pressure in a closed system. I added a pressure bleed-off to the vent line with a needle valve so I can tune tank pressure and it will also drop with RPM, though with some delay... I see a lot less gunk coming from my crank case pressure tap though. Does running it without the check valve impact the engines operation? It seems like screwing with the crank pressures could impact the tune - but I guess if it is predictable and consistent....

I have another solenoid sleeve on the printer now and will hopefully test again this afternoon on the way home from work. I plan to play with the .15 LA a bit more before I throw in the towel and focus the MVVS. I have a question about solenoid PWM frequency. Does anybody know what the frequency is for operating in Stihl chainsaws? I tried to set mine to 5hz but after seeing Raleighcopters post about the 8bit timer (Edit: 8 bit divider), and the fact that I "borrowed" his timer setup for my sketch - I was also running at ~37hz. I fixed my sketch to run at 5hz for this attempt just to balance solenoid life with reasonable throttle transition rate - but as I am typing this I think I will set it back to 37 hz since I got that somewhat working yesterday and I am changing other variables.

Lastly, I think to make this work well with the .15LA it needs a new carb. I dont think a metal carb will be a huge issue as I am not yet observing any vapor lock or bubbles in the fuel line from heat transfer - but the stock jet is massive. I printed that jet in the above pic with a .3mm hole and thats as small as my ****ty printer can go. I have some .1mm drill bits on order and plan to machine a whole new carb or at least carb jet with .1 and .2mm outlet. Then I will need a very good fuel filter - but I suspect the remote needle valve configuration, garbage plastic parts and general sensitivity of a tiny engine (and my junk printed parts) are the biggest issue I am having with getting the .15 running right.

The ASP has a keyed propwasher, BUT, a collet is absolutely NOT a problem.

I tie my sensors around the front bearing with a tie-rap, which makes them easily adjustable, virtually over 270 degrees of range.
Now if you fit your engine the first time, do the following.
Make sure that with the piston in top the magnet is somewhere where it is convenient to place the Hall Sensor.
With a scribe or similar, LIGHTLY mark TDC against a recognizable point on the engine, like for example the moldline of casting the crankcase.
Measure the diameter of the propwasher (locally where you made the scribe mark) with a vernier gauge/vernier calliper, and multiply the measurement by 0,2419 (easiest to use the metric numbers).
Set your vernier caliper at the outcome of this computation, and lock the slide.
Now place one point of the caliper on the scribed mark, such that the other point is WITH the direction of rotation. Mark this point also with a scribe, also lightly. (lightly so you can polish them out afterwards)
Now these marks are, if you worked accurately, exactly 28 degrees apart.
Turn the 2nd mark to the reference point, and set the sensor according to the instruction leaflet of your ignitionsystem.

NOW comes the important part: turn the engine to EITHER top or bottom dead center, whichever you find is easiest to replicate (personally, I take out the plug, and push the piston down with a suitable wooden dowel, the most secure way to find BDC).
On the propdriver, mark where the hallsensor is, place a permanent scribe mark at either side and also permanently shade the area between these two marks.
Polish out the TDC and 28 degree marks, because they have become useless.

Whenever you need to disassemble your engine, timing is easily restored by fitting the propdriver in "approximately the right position", then turn the crank to BDC (or TDC if you opted for that position), ans slide the sensor in place. Done...

I was told by a licensed Stihl Mechanic that it was 30 Hz.
I experimented with 15 to 35 Hz, and I noticed that below 25 Hz, low speed adjustability began to suffer appreciably.At low RPM, an "interference oscillation"between engine and solenoid frequency became increasingly noticable, and at 15 Hz, lowest achievable idle RPM went up to values I did not like.
I could not detect real differences above 30 Hz, but had the vague impression that above 30Hz the range of the valve got "smaller". BIt hard to explain. Still the same flow capacity, but the upper and lower 10% of the range behaved a bit strange. Nothing I could really put my finger on though.

My drivers run at or very close to 30Hz.

Hang on... you have a remote needle config?
What is the material? I have had horrible issues with the remote needle of my first conversion (OS FXi 46) being made out of metal, and heat transfer from crankcase to Remote NVA caused irresolvable issues with vapour... When I decoupled that NVA thermally from the engine, those vapour issues were gone.

Don't know if this helps, but...

Do you have overview pics of the entire install, showing some detail?

This is a fine way to do it. I considered doing some kind of indexing marks but couldn't muster the mental horsepower to derive it myself. This is straight forward and simple. Thanks again for that tidbit. I was seriously bummed when I saw the collet, thinking I would have to do the 46AX on spark instead of the MVVS like I wanted...

I am glad I asked, this saves me the trouble of that work. 30hz it is!

The .15LA originally used a remote needle valve with a plastic backplate. I removed this right off due to severe air intrusion into the line. I replaced the leaking plastic backplate with a resin-printed backplate incorporating a crank case pressure tap and O-ring to seal. So far it is working but I plan to delete the crank pressure tap and just use muffler pressure as you suggested.

I could give you some pics but my setup changes every time I go try it and learn something. I will take more pics of todays run and update later but now there is no needle valve at all and I am printing inserts to slip into the carburetor. Those inserts go where the original plastic fuel jet and fuel line nipple went (it is just pressed into the side of the aluminum carb body). My printed part has a jet diameter of about 0.3mm where the original was around 1mm (I dont have pin gauges to actually measure this). I know this needs to be smaller so hopefully a machined jet in the future will help here.

i see air bubbles in the line from the tank to the solenoid. are you sure air isn't getting in at the tank plumbing?

I wish I could isolate it back to this (it would be simple to fix) but I suspect those bubbles are just there because I had just tipped the test stand back to dump fuel from a flooded engine. I am not sure exactly how they got there in that pic but they arent there during normal operation.

Learning and subsequently changing things is the way to go... But... Document yourself to keep track of what you did. Sometimes one change don't work, the next one also doesnt, the third one did a bit, then a combo between 1st and 3rd (or any other random combo of on their own seemingly unsuccesful "solutions") sometimes turns out to hit the sweet spot...

I see you connected the muffler pressure line out of Tygon straight to the nipple. That will not hold, Tygon cannot handle real heat. You can shorten the line, but at some point it is too short.
A very good solution is 1" of silicon tubing on the nipple, 1/2~3/4" of brass tubing 1/8", then the Tygon. The Silicon is sacrificial but will hold several years without issues.