Thank you The Ridge for the awesome wallet and for supporting the channel! Check them out here: ridge.com/2strokestuffing Use code 2STROKESTUFFING for 10% off!
Thank you for posting such amazing content. I've watched your whole channel in the past couple of months and it's awesome now that you're going full time with it.
Wish I had your code when I bought mine, lol. Still though, bought one a little more than a year ago for a birthday present to myself, and it's awesome. I'd never go back to a normal wallet.
One thing I learned back when I was working with scales and load cells is that you always want some soft rubber or the like between the load cell and whatever you masseur. The metal in the load cell is basicly a very good spring, and it will oscillate like crazy if you don't have something that absorbed the vibrations. (And turn them in to heat.). It makes a HUGE difference.
Exactly i was thinking the same thing. Adding a shock absorber or such between the loadcell and engine mount point should help dampen down the signal. It wont affect total load seen as it will always be the same, but the large sudden spikes should be dampened down. Im currently building a water brake dyno which has a natural tendency to dampen shock waves just due to the tendency of the water, but typically the load cell is attached to the "torque load" (brake) which usually sees much less harsher erratic vibrations, then being attached directly to the engine. But hey, its all in the fun of experimentation
I've been a motorcycle enthusiast and video professional for over 50 years. Your content is amazing. Being a 2T addict, I emotionally share in each success and/or failure. The following may help with your video editing/color correction: The the difference in the digital algorithm between 8 bit color and 10 bit is geometric. The difference for what you are doing is not worth the processing load you are placing on your computer during 10 bit editing. Also, do not shoot in RAW mode. Use a 4:2:2 CODEC with mild compression, such as Apple Pro Rez. For color correction Black Magic Designs makes the de-facto standard software and offers it for free inside their "Resolve" editing software. If Pro Rez is not available in Resolve, use a CODEC with low compression ratios. Avoid RAW and Uncompressed. Those formats are more valuable when creating visual effects or presentations made for movie theaters. For what you are doing, the audience will never see the difference and your editing time will be greatly reduced. There were a couple of scenes where you got very close to a film-like "look". Very nice and aesthetically pleasing.
@@JamesHryb Good point about You Tube compression, James. It's mp4, which is very efficient, but 10 bit is a bit overkill for TSS style video. He showed a flash frame of the camera-original which made we think he also shot in RAW. More is not always more in digital video. Thanks.
In the beginning of the 1960's, 50cc two-stroke world championship grand prix bikes were making nearly ten horse. They didn't make 20hp+ until the early 1980's, thanks to rotary induction.
@@malcrandall1309 I think my favorite rider, funny enough, is Aalt Toersen. He still rides, tunes, builds.. Still committed to Florett engines, and staying true to Van Veen.
Every video I'm more and more impressed with your progress and determination. Your video releases are easily one of my most anticipated notifications on youtube. Keep up the great work, sir!
Two things to thank about. Nitro has a very slow flame propagation. Top fuel dragsters run up to about 54 Deg spark lead. You can also try a simple RC engine glo plug. They make a adapter that will let you do this. Next,,,a carburetor is not a reliable delivery device when it comes to Nitro. You need a fuel injection method of some kind. I don`t know if you know this, but you can do away with the pulse line on your pumper carb, and just use a electric fuel pump in the fuel line. Block off the pulse line altogether.
Awesome work! I really like how you are just saying exactly what is happening without a filter and talking about ideas that other folks have given you. Nothing is off the table and you are willing to try anything. This is refreshing to see.
Hey man, I really like your work and I would like to suggest you something: -consider creating an external management panel with all the output data, digital knobs for handling stuff like ignition time and kill switch -connect a servo for throttle, this will allow ECU to handle engine RPMs and will make it easy to troubleshoot and calibrate the engine -and for the love of OHS (Health and Safety), PLEASE cover this chain :D If you need help with anything above contact me, I will try to help :)
i dont ride, i dont 2stroke, i dont even have a reason to learn anything 2stroke. i 4x4 on a mountain, play with fdm printers, and work on 4stroke engines.... i have no reason to be in this room!! but hell, i really like watching this guy!
It’s all coming together! Hard work really pays off! I remember When you were still casting the cylinder at home. And now you are giving us dyno action! Keep up the work!
I was really sitting here thinking you were playing at the end thinking there was more. I think the raccoons have crawled under your hat. Keep up the good work nice motor!!
Your shop was always organized but since you started this full time I can see a big difference in how much more organized and well put together everything is. Well done sir your living your dream!!! Hell yeah brother!
We use these carbs in Karting.. Anything over 2 turns.. and its FULLY OPEN... like any tap.. Lower the pop off to around 8psi... BUT.. you need to set the needle and seat.. PERFECTLY... adjust the needle fork to make sure you get consistent POP OFF.. without leak down.. The pulse is VERY important.. get pulse from the crank case.. ( but the supercharger may be a problem... as its over atmosphere pressure ) ... ALSO .. you can drill the jets.. BUT that is drastic ..
Haven't finished the video but maybe a boost referenced fuel pump that will raise the fuel 1psi per every psi of boost. Common thing with forced induction cars to counteract the pressure the fuel injectors see inside the cylinder.
Common with forced induction where you have injectors that is. If you suck thru a carb, all it will see is a higher gas velocity/lower pressure. My best bet would be to go crazy with the spark energy to remove one variable from the problem. Mounting a MAF sensor and logging the output would probably help to.
@@zlaya84 good point. I also agree that having more data to go by like a MAF and a wideband for A/F ratio would be a great addition. With that kind of data it would be cool to go ahead and switch over to fuel injection and be able to tune this thing to perfection.
@@zlaya84 or maybe even mechanical fuel injection... would be great for keeping the cylinder lubricated when off the throttle but really I have no idea if that would work on an engine like this.
float carb would be annoying when running it on the road, but a fuel pump would solve your problem (or make it worse by adding something else to the equation...)
@@afflickracing1705 oh ok. I know way too much people who blew up bikes because of leaning out. I hadnt finished the video though and it seems to work out for him so good
It's not presssure its volume at issue. A fuel pump wont fix it per se. Whatever is limiting the volume of fuel to the carb, outside of pressure, is the issue. Increase the volume, at the same or lower pressure even likely and wham bam we are off to the races!
One thing I learned from watching some dyno tests is that ignition timing is what gives the power once all of the fuel/air mix needs have been met, too much fuel quenches the plug. Initial advance in timing from where you are now will give a quick increase and then a diminishing return, there is an issue with flame front speed based on fuel/air ratio and with your high revs this is important. On my microlight Subaru engine timing up to 30 deg gives good power, 33 deg a little more but not worth it, be conservative for reliability. Good luck, interesting work, appreciate your efforts!
11:40 you can see on the joint of that pressure thing under the motor, that it is not probably sealing... some wierd white stuff is going through seating surfaces
Hey! I might know the problem of the dyno( sorry for mistakes, 2. Language) so because the motor pulls the chain off- center( rellative to the load cell pivot point) the radius of the gear, and the distance of the chain from the pivot point make a gear-ratio diferent from1/1 . This way you don't meassure the torpue of the motor, but the torpue generated by the chain-pulling force on the pivot point.
maybe hooking a running motor with known HP rating would help validate the dyno. You're running with too much unknown variables IMHO. Great job as always.
I've been watching your channel for a while now and I just wanted to say, good job! I've always been a man who loves to tinker with all manner of different things and have rode 2 stroke dirtbikes for a long time now. It's always been interesting to see your different engine designs and just watch your journey towards the most powerful 50cc 2t engine. If and when you ever make it back to the Bonneville salt flats let us know ahead of time. I live pretty close to the salt flats and think it would be cool to come show our support in your attempts at the record.
Tuning the pid values is a good idea. I think one problem may be the rotation sensor reading from the crank. One engine revolution is so short in time that it causes large fluctuations in the rotation speed data. This probably causes the brake control loop to overreact and apply too much brake. For the engine bogging problem I suggest applying dyno load manually and seeing if the bogging is really caused by engine problem under load or just too much brake. I don't think the load cell has an effect on the dyno ramp control. At least in my diy brake controller it only measures the roller acceleration and reacts based on that to adjust the load. Still need to test it though if it even works.
To check torque sensor calibration you could safely lock the engine (e.g. theaded alu bar though spark plug hole to stop the piston halfway up the cylinder) and then use a torque wrench on the engine drive sprocket.
I think you need to separate your issues. Test the dyno with a reliable stead state engine, either a basic bike engine (the Derbi?) or something like a Honda GX engine, reliable, quiet and easy to run all day. Once dyno is setup you should play with the unknown “brute force” engine, in good faith that your dyno is working perfectly 👍
As an engine engineer, I recommend a dyno that can hold constant RPM. Then you can figure out what your engine is doing. Also, I recommend looking at measuring CO and if you're lucky, O2 in the exhaust gas. Lambda sensor won't do it for measuring AFR. Then you can really start to diagnose your engine. Also, use a known good engine for tuning your dyno.
On turboed n/a cars, ignition plug gap is important. Make sur to have a short good gap :) have to do all my gaps at each turbo build ... Or i have no ignition when boost comes up, hello from france
Ok - now at 5:23. 8 bit with more contrast - far richer than with lower contrast. Weirdly - the background cans and milling head (when you are at your desk) are also in sharper focus at this higher contrast setting. Higher contrast is definitely richer/ better looking to my eye
A hot coil setup could definitely help. I've had that problem with stock COPs on higher boost turbo engines and had to upgrade the COP and use a different plug.
Looking at your vidéo I think the chain needs some sort of tensioner on the unloaded side. The chain vibration creates a pulsating load that pollute load cell readings....
All kinds of vibration, Imbalance, forces will be present in your load cell data with that setup. Along with your chain reaction force as you have 2 load paths present. You'll want to check for electromagnetic interference from your ignition as that's a strain base load cell. When you calibrate that load cell you need the perpendicular distance to the direction of the force not to the bottom pivot on the load cell. draw a line thru the 2 pivots on the load cell the distance perpendicular to that line and the center of your engine pivot is your moment arm length it doesn't look like much error but hard to tell from pictures.
Top Video!!!! Probable , The fuel level dropped below the Main jet pickup. Suggestion up size the passages and inlet valve. place a smaller load cell at a larger dia, like 1m with a dampener at .5m.
If your going to use PWK carb you need to check if high flow shut off valve is available? There are things that can be done to stock shut off to increase flow but much of the modifications have not been available in print since the 1970's.
Awesome work mate if it keeps dieing in revs full throttle go bigger jet or 35mm carb , full boost it's running out of gas full revs. As I've said before do a plug chop , only way to see colour of plug full throttle especially when advancing timing you don't won't a lean out full throttle. Keep it up.
Just a suggestion; it dawned on me tonight that your operations may be magnitudes easier if you mount your engine and brake ON TOP of your 'project beam' to minimize any '2strokestuffing' around on the ground (more comfort, less whiskey) while comfortably standing up to work LOL. Great vids, trial and error, and engineering shenanigans man!
I hesitate to suggest non-engine-related work, but your process is so relentlessly iterative that I'm sure you can make much faster progress if you improve your workflow. To wit: why not raise the jackshaft sprocket more above the retarder? That raises the engine enough that 1.you can make changes from a wheely stool instead of on your knees, which is 2. more comfortable, 3.gives you better access and visibility to all components, 4.makes it easier to position yourself so you're not blocking the camera as much. 5.Allows you to shorten the longitudinal extrusion, reducing the overall footprint and recapturing shop space. 6.Tightens the layout giving you more space in the shot for overlays ( or dramatic visual expression, Sir David Lean ) 7.Potentially shortens the chain. 8.Makes a vertical ( then across the ceiling and back down ) exhaust duct feasible removing the trip hazard on the floor. And, on that subject, some kind of smaller diameter, preferably fireproof rigid metal, initial duct with appropriately spaced transitions can almost certainly improve draft velocity and do a better job of keeping the smoke in the neighbor's yard, where it belongs.
Getting closer to the magical formula! What is the maximum boost pressure you're running? Advancing the ignition timing sounds like a good plan to me. Your videos are great food for thought, and the music is perfect, from a long time sub here
Increese your timing advance to 20-25 degrees at full throttle. Example outbord 2-t motors change timin manually from 6deg at idle to 23 deg full throttle
My dad used to be into 2 stroke model airplanes, he told me once that if you had something off balance and had a vibration just right, the fuel would froth up and the engine would lean out. idk if that pertains to your situation but worth a thought maybe.
Needs PID tuning, and check for rpm spikes by graphing rpm against time, make sure to ground the eddy brake housing to stop interference when the load is applied, there is an option to set a percentage of load to start with this can help with smooth load application try 10%, you can add a plugin to show what the PID is doing with live graphs on the run screen for kp ki kd
Nice work as always! I have been wondering for a while now if it would be worth deactivating some of the coils in the retarder so it's not so aggressive with these smaller loads.
I agree, I think the retarder is way too powerful for the engine he's using, which means the control loop gain is very high, which invites instability.
Also, unless you have a very powerful ignition system, reducing the spark plug gap is something often needed on a boosted engine. What is the gap set at now?
Really depends on how much boost. If it's a bar or under you don't need too powerful of ignition or tight gap. I'm going to bet he is running very little boost, running it out of fuel under high boost probably would have trashed it already. Been there done that.
I don't think it's the boost Blowing out the spark but idk this is uncharted territory... I think if the ignition is being blown out I think it's all the fuel.. it seems to me that the float carb is dead rich.. I think this setup can definitely achieve 10 pounds of boost tho (.7 bar) and that can blow out a spark if the gap is too big.. also very little is known about how an engine setup like this will respond to boost, timing, fueling, everything... there's exhaust valve timing that has to be worked out... You can't just look this stuff up lol and he's going to have to burn a piston or melt a plug to know how lean is too lean no o2 sensor will give you accurate afr because of the way the blower blows raw fuel through the thing all the time which will also mess with egt sensors too
I agree, we run 24psi and some ignitions just cannot fire under that cylinder pressure. You can close the gap, but that’s a Band-Aid solution not a fix. We run a Dyna S on our 100cc, 50% nitro and almost 2-1 fuel air and it fires great. We have even gone to a ion battery so the coils gets more power.
Nice progress. I like your experiments with cinematic style. Roughly how many horsepower are you expecting to achieve with this design? It is hard to tell how close to achieving your gol you are?
@@RotarySMP he is trying to push it to the limit of what is possible, with his transfer ports set the way they are, his supercharger, and the fact that he is running nitro rather than gas it has a higher limit than what a different 50cc running gasoline could do.
when we dial in our dyno we did it up in increments. like start at zero then go 1 lb then 2 lb then 3 then 4 then 5. then do the same on the way down. also you could use a oil trq system with popit bypass. we also have a very useless electric trq system that i would totally stay away from. also the problem could be in your data acquisition you might have to damper the reading out like a filter
Maybe an oversized float bowl, or an exhaust pulse to fuel tank setup, which may be able to overcome the float needle pressure. A flame arrestor and variable pulse valve would help control things a bit.
You will get there after exploring all other options, it’s good to watch and asap set up some straight thought mufflers when you get a base line one constant power readings. Because once your hearing goes it does not come back.
Hi, what may help to reduce fluctuations in the readings from the loadcell might be to put a chain tensioner on the large final drive chain to the retarder. I notice it is flapping and bouncing about during you runs. Keep up the good work.
You're all over it the pid loop is very important for steady state and ramping the engine. I've found i need a fair bit of P.. A little bit of I... And no D at all on my chassis dyno. And derivative at all and it would over and under shoot the target
I am a marine mechanic and i tune a lot of race outboards for customers. I completely agree with the 15degrees of timing being a huge issue. The engine will start to drown in fuel at higher rpms. A lot of the smaller displacement race engines are running 30-45 degrees of timing at 9000-10000rpm. Is the engine running with fixed timing or is there an advance in your system? Also the spark plug might need to go up or down in heat range once its running better.
The Maha lp13 has a Telma retarder. The Axle holds the weight of the retarder and the load cell measures how much rotational force the Telma retarder is holding back. Let me know if you need some pictures of the setup.
With the load cell you have now leaving it in its current position is probably the best option if a rubber damper is fitted between the load cell and the engine cradle, making a longer arm will increase the amount of error reading as a smaller part of the range is used. With a longer arm a load cell with a smaller range is needed, like 25 or 50N. But the longer arm could be used as a damper in itself, when it has some weight at the load cell end, the inertia will dampen fast occurring changes in load like those vibrations, just be sure to zero out the static load well.
Awesome! Progress! I was thinking in the last video that the retarder seems to come in too hard and the PID values might need looking at. Halving the voltage seemed promising and meant that it used more than 1 or 2% of the retarder range. Then you mentioned it doesn't have PID control? Definitely something to look into. Are you able to hook up an electric motor in place of the engine to get a baseline for the power input and calibrate the software from that? Then you can also switch the motor on from a dead stop and have the retarder ramp it up to a certain rpm over a certain time and hold it at whatever rpm you like, which is what it should do with the engine. Not sure if you factored it in previously, but the calibration weight hanging from the load cell bolt isn't accurate because of the angle (45 degrees) of the moment arm. The bolt @ 14cm will be much better as it's nearly horizontal with the engine pivot. Keep up the good work.
Go for a bigger load cell and mount it of the retarder as you stated. Also what is the resolution and sample rate of the software. You need at least x5 the rpm sample rate so at least 1kHz. You could run an accumlator for the pressure for the carb, or run it off the crankcase pressure. The accumulator would just be a bottle and then run that to the carb if you're worried about exhaust gases. Just make the lines short to reduce latency.
I think your on the money with the the pid controll not being right less alot less proportional to start with until you can get it to hold the engine at a desired rpm when the load is applied smoothly then start on the intergral so it doesn't overshoot target rpm when hit hard
I don't know what percentage you're running but you'll need a bigger float bowl and more than likely needle and seat area. Straight nitro is around 1.7/1 AFR and alcohol is in the 6/1 range so you'll need more reservoir for sure. As far as ignition is concerned, nitro burns slow and will need more timing more than likely.
Methanol burns slowly compared to petrol so will tolerate more advance, and needs to be jetted much richer. I would have thought that 15 degrees is way under advanced and risks running hot. I think this engine has lots more to give and am very impressed with what you are doing. Can't wait to see what you can get out of it.
Forget the carb, use electronic fuel injection. Get megasquirt or something like that and an inlet manifold from a fuel-injected motorbike, strip away the extra throttle bodies if you get a 4-cyl or 2-cyl one.
Log format video gives you more leeway in controlling contrast in "badly" lit scenes but that's about it. Would love to see rpm sweeping using something like an arduino hooked up to a relay to PWM control the retarder to a somewhat stable RPM range (could even make a sweep protocol with that sort of system, especially with a servo hooked to the throttle).
I'm experimenting with log cause almost every shot involves shiny aluminium, hard to expose properly without blown highlights. Not that it's of much significance, but I can't help it. And a complicated process makes it more interesting.
The dyno is hard to calibrate, everyone I know ended up buying a motorcycle dyno. Same with your timing issues, you can put a fuel tech or megasquirt EFI on it and get actual control and Afr.
Load cell calibration from centre of pivot or centre of crank? You may have some weird dynamics going on from remounting the engine in the cradle for the PIP. Also, I think dampening the load cell from that angry little engine is a good idea. Hope this helps. Great content! 👍
i feel like you're probably getting way too much oscillation from the engine. maybe stiff rubber bushings that take up the slop but pad both sides of the load cell? could also put a long bar on the engine to dampen the rotational acceleration (but then you get into wave forms in the bar). also, under boost do you have enough voltage to spark plug so that it isn't being blown out? could decrease gap to test it.
Thank you The Ridge for the awesome wallet and for supporting the channel! Check them out here: ridge.com/2strokestuffing
Use code 2STROKESTUFFING for 10% off!
Thank you for posting such amazing content. I've watched your whole channel in the past couple of months and it's awesome now that you're going full time with it.
Wish I had your code when I bought mine, lol. Still though, bought one a little more than a year ago for a birthday present to myself, and it's awesome. I'd never go back to a normal wallet.
if you send the excess air between the compressor and the carburator, that engine will consume waaaaaay less fuel
Are those freedom dollars in your Ridge wallet?
@@mrpresidentm1 in norway is hard to have freedom dollars
One thing I learned back when I was working with scales and load cells is that you always want some soft rubber or the like between the load cell and whatever you masseur.
The metal in the load cell is basicly a very good spring, and it will oscillate like crazy if you don't have something that absorbed the vibrations. (And turn them in to heat.). It makes a HUGE difference.
Like this so he will see this
Exactly i was thinking the same thing. Adding a shock absorber or such between the loadcell and engine mount point should help dampen down the signal. It wont affect total load seen as it will always be the same, but the large sudden spikes should be dampened down.
Im currently building a water brake dyno which has a natural tendency to dampen shock waves just due to the tendency of the water, but typically the load cell is attached to the "torque load" (brake) which usually sees much less harsher erratic vibrations, then being attached directly to the engine.
But hey, its all in the fun of experimentation
“This Is The Way”
This is a much better solution than mounting a damper parallel to the load cell, as that will cause an additional load reading.
I've been a motorcycle enthusiast and video professional for over 50 years. Your content is amazing. Being a 2T addict, I emotionally share in each success and/or failure. The following may help with your video editing/color correction:
The the difference in the digital algorithm between 8 bit color and 10 bit is geometric. The difference for what you are doing is not worth the processing load you are placing on your computer during 10 bit editing. Also, do not shoot in RAW mode. Use a 4:2:2 CODEC with mild compression, such as Apple Pro Rez. For color correction Black Magic Designs makes the de-facto standard software and offers it for free inside their "Resolve" editing software. If Pro Rez is not available in Resolve, use a CODEC with low compression ratios. Avoid RAW and Uncompressed. Those formats are more valuable when creating visual effects or presentations made for movie theaters. For what you are doing, the audience will never see the difference and your editing time will be greatly reduced. There were a couple of scenes where you got very close to a film-like "look". Very nice and aesthetically pleasing.
I could not tell the difference between the 8 or 10 bit in the video. pretty sure TH-cam compression killed any enhancement that it would have added.
@@JamesHryb Good point about You Tube compression, James. It's mp4, which is very efficient, but 10 bit is a bit overkill for TSS style video. He showed a flash frame of the camera-original which made we think he also shot in RAW. More is not always more in digital video. Thanks.
Germany?
@@genau6872 If you're asking where I'm from, its the middle of the USA.
Any video that has a running engine, and nothing destroys itself is a good video!
Keep it up, dude!
"Any video [...] is a good video!" - here, I fixed it
In the beginning of the 1960's, 50cc two-stroke world championship grand prix bikes were making nearly ten horse. They didn't make 20hp+ until the early 1980's, thanks to rotary induction.
I can remember seeing Henk van Kessel riding the van Veen Kreidler at Spa years ago. A tall man on such a tiny bike.
@@malcrandall1309 One of the great riders!
@@malcrandall1309 I think my favorite rider, funny enough, is Aalt Toersen. He still rides, tunes, builds.. Still committed to Florett engines, and staying true to Van Veen.
Every video I'm more and more impressed with your progress and determination. Your video releases are easily one of my most anticipated notifications on youtube. Keep up the great work, sir!
Two things to thank about. Nitro has a very slow flame propagation. Top fuel dragsters run up to about 54 Deg spark lead. You can also try a simple RC engine glo plug. They make a adapter that will let you do this. Next,,,a carburetor is not a reliable delivery device when it comes to Nitro. You need a fuel injection method of some kind. I don`t know if you know this, but you can do away with the pulse line on your pumper carb, and just use a electric fuel pump in the fuel line. Block off the pulse line altogether.
Awesome work! I really like how you are just saying exactly what is happening without a filter and talking about ideas that other folks have given you. Nothing is off the table and you are willing to try anything. This is refreshing to see.
Hey man, I really like your work and I would like to suggest you something:
-consider creating an external management panel with all the output data, digital knobs for handling stuff like ignition time and kill switch
-connect a servo for throttle, this will allow ECU to handle engine RPMs and will make it easy to troubleshoot and calibrate the engine
-and for the love of OHS (Health and Safety), PLEASE cover this chain :D
If you need help with anything above contact me, I will try to help :)
Polska ❤️ has a really good idea! Which brings me to a thought; outsource stuff like programming in this case so you can concentrate on reving!
Oh God the engineers have arrived, hide the tools! lololol :{)=
i dont ride, i dont 2stroke, i dont even have a reason to learn anything 2stroke.
i 4x4 on a mountain, play with fdm printers, and work on 4stroke engines....
i have no reason to be in this room!!
but hell, i really like watching this guy!
It’s all coming together! Hard work really pays off! I remember When you were still casting the cylinder at home. And now you are giving us dyno action! Keep up the work!
I was really sitting here thinking you were playing at the end thinking there was more. I think the raccoons have crawled under your hat. Keep up the good work nice motor!!
The new camera looks great, little brightness couldn't hurt though!
Exactly
Jeg er SINDSYG glad for dit video format OG alle de andre ting du gør "uden for emnet" !! De løfter to-takter indholdet til HIMMELSKE højder
Håller med! Alla borde använda det här formatet på yt! (:
I have had the same fuel issue with those Tillitson carbs, there is a transfer port you can drill out
It’s upside down. Methanol tillotson has to have the pumps and fuel feed on top.
Your shop was always organized but since you started this full time I can see a big difference in how much more organized and well put together everything is. Well done sir your living your dream!!! Hell yeah brother!
We use these carbs in Karting.. Anything over 2 turns.. and its FULLY OPEN... like any tap.. Lower the pop off to around 8psi... BUT.. you need to set the needle and seat.. PERFECTLY... adjust the needle fork to make sure you get consistent POP OFF.. without leak down.. The pulse is VERY important.. get pulse from the crank case.. ( but the supercharger may be a problem... as its over atmosphere pressure ) ... ALSO .. you can drill the jets.. BUT that is drastic ..
Haven't finished the video but maybe a boost referenced fuel pump that will raise the fuel 1psi per every psi of boost. Common thing with forced induction cars to counteract the pressure the fuel injectors see inside the cylinder.
Common with forced induction where you have injectors that is. If you suck thru a carb, all it will see is a higher gas velocity/lower pressure.
My best bet would be to go crazy with the spark energy to remove one variable from the problem.
Mounting a MAF sensor and logging the output would probably help to.
No need, it's a draw through setup not blow through
@@zlaya84 good point. I also agree that having more data to go by like a MAF and a wideband for A/F ratio would be a great addition. With that kind of data it would be cool to go ahead and switch over to fuel injection and be able to tune this thing to perfection.
@@zlaya84 or maybe even mechanical fuel injection... would be great for keeping the cylinder lubricated when off the throttle but really I have no idea if that would work on an engine like this.
float carb would be annoying when running it on the road, but a fuel pump would solve your problem (or make it worse by adding something else to the equation...)
Float carb will work ok, we run huge amounts of fuel with nitro through one on our 100cc
@@afflickracing1705 oh ok. I know way too much people who blew up bikes because of leaning out. I hadnt finished the video though and it seems to work out for him so good
@@scottex. yes me as well, might have been bad setups. check out our big remote fuel bowl.
It's not presssure its volume at issue. A fuel pump wont fix it per se. Whatever is limiting the volume of fuel to the carb, outside of pressure, is the issue. Increase the volume, at the same or lower pressure even likely and wham bam we are off to the races!
I love all the stuff about cameras! The difference in video quality vs your older videos is incredible!
One thing I learned from watching some dyno tests is that ignition timing is what gives the power once all of the fuel/air mix needs have been met, too much fuel quenches the plug. Initial advance in timing from where you are now will give a quick increase and then a diminishing return, there is an issue with flame front speed based on fuel/air ratio and with your high revs this is important.
On my microlight Subaru engine timing up to 30 deg gives good power, 33 deg a little more but not worth it, be conservative for reliability.
Good luck, interesting work, appreciate your efforts!
I once commented about using the exhaust for the pulse, it was amazing i wast the only one thinking about it, love the vids, keep up the amazing work!
2t sidecars used exhaust pulses for their Tillotsons (sp?) Back then anyway.
There's some experienced discussion going on here. No Fake News!
11:40 you can see on the joint of that pressure thing under the motor, that it is not probably sealing... some wierd white stuff is going through seating surfaces
that part where you synced the engine revving with the background music was great!! My favorite part of this awesome video
Hey! I might know the problem of the dyno( sorry for mistakes, 2. Language) so because the motor pulls the chain off- center( rellative to the load cell pivot point) the radius of the gear, and the distance of the chain from the pivot point make a gear-ratio diferent from1/1 . This way you don't meassure the torpue of the motor, but the torpue generated by the chain-pulling force on the pivot point.
👍 yep. Need to fix that with a new engine cradle, or I might be able to correct for it in the software.
maybe hooking a running motor with known HP rating would help validate the dyno. You're running with too much unknown variables IMHO. Great job as always.
7:30 you can see The smoke stop during the pulls at the duct tape exhaust fan connection, aka lean or no sparky
I've been watching for a long time now and just want to say that it's great to see someone putting so much effort into 2 strokes. 👍
I've been watching your channel for a while now and I just wanted to say, good job!
I've always been a man who loves to tinker with all manner of different things and have rode 2 stroke dirtbikes for a long time now. It's always been interesting to see your different engine designs and just watch your journey towards the most powerful 50cc 2t engine.
If and when you ever make it back to the Bonneville salt flats let us know ahead of time. I live pretty close to the salt flats and think it would be cool to come show our support in your attempts at the record.
Tuning the pid values is a good idea. I think one problem may be the rotation sensor reading from the crank. One engine revolution is so short in time that it causes large fluctuations in the rotation speed data. This probably causes the brake control loop to overreact and apply too much brake. For the engine bogging problem I suggest applying dyno load manually and seeing if the bogging is really caused by engine problem under load or just too much brake.
I don't think the load cell has an effect on the dyno ramp control. At least in my diy brake controller it only measures the roller acceleration and reacts based on that to adjust the load. Still need to test it though if it even works.
Thanks for the tips! I can easily move the sensor back to the retarder, and some manual brake testing is on my menu tomorrow.
To check torque sensor calibration you could safely lock the engine (e.g. theaded alu bar though spark plug hole to stop the piston halfway up the cylinder) and then use a torque wrench on the engine drive sprocket.
That's a great idea!
Of all my subscribed channels this is the one i always watch first if there is something new. Dude you make me smile.
I think you need to separate your issues. Test the dyno with a reliable stead state engine, either a basic bike engine (the Derbi?) or something like a Honda GX engine, reliable, quiet and easy to run all day.
Once dyno is setup you should play with the unknown “brute force” engine, in good faith that your dyno is working perfectly 👍
👊🏼 yes reduce variables.
electric motor for smoothness?
As an engine engineer, I recommend a dyno that can hold constant RPM. Then you can figure out what your engine is doing. Also, I recommend looking at measuring CO and if you're lucky, O2 in the exhaust gas. Lambda sensor won't do it for measuring AFR. Then you can really start to diagnose your engine.
Also, use a known good engine for tuning your dyno.
On turboed n/a cars, ignition plug gap is important. Make sur to have a short good gap :) have to do all my gaps at each turbo build ... Or i have no ignition when boost comes up, hello from france
You want the biggest gap possible, you only shorten gap if ignition does not have enough power to fire.
Ok - now at 5:23. 8 bit with more contrast - far richer than with lower contrast. Weirdly - the background cans and milling head (when you are at your desk) are also in sharper focus at this higher contrast setting. Higher contrast is definitely richer/ better looking to my eye
A hot coil setup could definitely help. I've had that problem with stock COPs on higher boost turbo engines and had to upgrade the COP and use a different plug.
Microsquirt may be the answer your looking for on fueling. Thanks for the video.
I was thinking the same thing. Fuel injection may be key on getting this beast running correctly under boost.
Put a guard round that chain! Specially when adjusting the carb whilst drinking rum.
Looking at your vidéo I think the chain needs some sort of tensioner on the unloaded side. The chain vibration creates a pulsating load that pollute load cell readings....
All kinds of vibration, Imbalance, forces will be present in your load cell data with that setup. Along with your chain reaction force as you have 2 load paths present. You'll want to check for electromagnetic interference from your ignition as that's a strain base load cell. When you calibrate that load cell you need the perpendicular distance to the direction of the force not to the bottom pivot on the load cell. draw a line thru the 2 pivots on the load cell the distance perpendicular to that line and the center of your engine pivot is your moment arm length it doesn't look like much error but hard to tell from pictures.
don't stop these video, they are soo interesting to watch at the end of the day !!!
I can't wait to see some result
Top Video!!!!
Probable , The fuel level dropped below the Main jet pickup. Suggestion up size the passages and inlet valve.
place a smaller load cell at a larger dia, like 1m with a dampener at .5m.
As long as the tank is above the carb it should work even without the pump side connected at all.
If your going to use PWK carb you need to check if high flow shut off valve is available?
There are things that can be done to stock shut off to increase flow but much of the modifications have not been available in print since the 1970's.
Awesome work mate if it keeps dieing in revs full throttle go bigger jet or 35mm carb , full boost it's running out of gas full revs.
As I've said before do a plug chop , only way to see colour of plug full throttle especially when advancing timing you don't won't a lean out full throttle.
Keep it up.
Just a suggestion; it dawned on me tonight that your operations may be magnitudes easier if you mount your engine and brake ON TOP of your 'project beam' to minimize any '2strokestuffing' around on the ground (more comfort, less whiskey) while comfortably standing up to work LOL. Great vids, trial and error, and engineering shenanigans man!
Maybe you can use a Air to fuel ratio meter in the exhaust to help tuning? you can buy them cheap on ALI with a meter attached
Inertia you can overcome by water turbo , use waiste gate etc, go forward
Duncan racing high flow Carburetor bowl. Made for the PK corporator you can buy right after website
Love your work, I always get to the end either laughing or at least smiling, admire your persistence.
I hesitate to suggest non-engine-related work, but your process is so relentlessly iterative that I'm sure you can make much faster progress if you improve your workflow. To wit: why not raise the jackshaft sprocket more above the retarder? That raises the engine enough that 1.you can make changes from a wheely stool instead of on your knees, which is 2. more comfortable, 3.gives you better access and visibility to all components, 4.makes it easier to position yourself so you're not blocking the camera as much. 5.Allows you to shorten the longitudinal extrusion, reducing the overall footprint and recapturing shop space. 6.Tightens the layout giving you more space in the shot for overlays ( or dramatic visual expression, Sir David Lean ) 7.Potentially shortens the chain. 8.Makes a vertical ( then across the ceiling and back down ) exhaust duct feasible removing the trip hazard on the floor. And, on that subject, some kind of smaller diameter, preferably fireproof rigid metal, initial duct with appropriately spaced transitions can almost certainly improve draft velocity and do a better job of keeping the smoke in the neighbor's yard, where it belongs.
All very good suggestions!
Love the jazzy chords and your work Thanks a lot !!
Getting closer to the magical formula! What is the maximum boost pressure you're running? Advancing the ignition timing sounds like a good plan to me. Your videos are great food for thought, and the music is perfect, from a long time sub here
Increese your timing advance to 20-25 degrees at full throttle. Example outbord 2-t motors change timin manually from 6deg at idle to 23 deg full throttle
My dad used to be into 2 stroke model airplanes, he told me once that if you had something off balance and had a vibration just right, the fuel would froth up and the engine would lean out. idk if that pertains to your situation but worth a thought maybe.
Needs PID tuning, and check for rpm spikes by graphing rpm against time, make sure to ground the eddy brake housing to stop interference when the load is applied, there is an option to set a percentage of load to start with this can help with smooth load application try 10%, you can add a plugin to show what the PID is doing with live graphs on the run screen for kp ki kd
I think it's been said already, but I really dig the glasses. Also, hearing it run pretty smoothly like this was very nice to hear.
Nice work as always! I have been wondering for a while now if it would be worth deactivating some of the coils in the retarder so it's not so aggressive with these smaller loads.
I agree, I think the retarder is way too powerful for the engine he's using, which means the control loop gain is very high, which invites instability.
Also, unless you have a very powerful ignition system, reducing the spark plug gap is something often needed on a boosted engine. What is the gap set at now?
I bet you're onto something here.
Really depends on how much boost. If it's a bar or under you don't need too powerful of ignition or tight gap. I'm going to bet he is running very little boost, running it out of fuel under high boost probably would have trashed it already. Been there done that.
Me too.
I don't think it's the boost Blowing out the spark but idk this is uncharted territory... I think if the ignition is being blown out I think it's all the fuel.. it seems to me that the float carb is dead rich.. I think this setup can definitely achieve 10 pounds of boost tho (.7 bar) and that can blow out a spark if the gap is too big.. also very little is known about how an engine setup like this will respond to boost, timing, fueling, everything... there's exhaust valve timing that has to be worked out... You can't just look this stuff up lol and he's going to have to burn a piston or melt a plug to know how lean is too lean no o2 sensor will give you accurate afr because of the way the blower blows raw fuel through the thing all the time which will also mess with egt sensors too
I agree, we run 24psi and some ignitions just cannot fire under that cylinder pressure. You can close the gap, but that’s a Band-Aid solution not a fix. We run a Dyna S on our 100cc, 50% nitro and almost 2-1 fuel air and it fires great. We have even gone to a ion battery so the coils gets more power.
Maybe a higher power ignition coil could help with igniting the mixture. The IGN1A is a popular one, not cheap, but you only need one at least!
I know nothing about A dyno but was wondering could you use a 4 horse power output electric motor to test Dyno accuracy???
Thats works really fine If it is IE3 grading power loss is guaranteed to be under 5%. so 4kw +/-5%
A 4 pole small motor of 2 hp operating at 50 Hz:
@@RonnyT89 sorry mate but was u speaking English there or am I really that clueless about how a dyno works 😂
Yeah I read it again ITS ME 🤣✌️🇬🇧
Nice progress. I like your experiments with cinematic style.
Roughly how many horsepower are you expecting to achieve with this design? It is hard to tell how close to achieving your gol you are?
Thanks! Anything below 30hp and it would be pointless. Needs more boost.
@@2STROKESTUFFING 600Hp/L That is quite a bit :)
Lovely numbers❤️❤️❤️👍👍👍
@@RotarySMP he is trying to push it to the limit of what is possible, with his transfer ports set the way they are, his supercharger, and the fact that he is running nitro rather than gas it has a higher limit than what a different 50cc running gasoline could do.
when we dial in our dyno we did it up in increments. like start at zero then go 1 lb then 2 lb then 3 then 4 then 5. then do the same on the way down. also you could use a oil trq system with popit bypass. we also have a very useless electric trq system that i would totally stay away from.
also the problem could be in your data acquisition you might have to damper the reading out like a filter
One of the 1 or 2 channels that I never skip the adds even if it part of the post... you rock Bro!
Maybe an oversized float bowl, or an exhaust pulse to fuel tank setup, which may be able to overcome the float needle pressure. A flame arrestor and variable pulse valve would help control things a bit.
Tighten your gap on the plug. Bigger coil. The compression makes it hard to bridge the gap.
You will get there after exploring all other options, it’s good to watch and asap set up some straight thought mufflers when you get a base line one constant power readings. Because once your hearing goes it does not come back.
You could use good quality calibrated torque wrench to calibrate the load cell. That way you'll not have reduced sources of error.
I mean you would reduce error sources... Btw. Great content, keep up the good work!
Hi, what may help to reduce fluctuations in the readings from the loadcell might be to put a chain tensioner on the large final drive chain to the retarder. I notice it is flapping and bouncing about during you runs. Keep up the good work.
You're all over it the pid loop is very important for steady state and ramping the engine. I've found i need a fair bit of P.. A little bit of I... And no D at all on my chassis dyno. And derivative at all and it would over and under shoot the target
I am a marine mechanic and i tune a lot of race outboards for customers. I completely agree with the 15degrees of timing being a huge issue. The engine will start to drown in fuel at higher rpms. A lot of the smaller displacement race engines are running 30-45 degrees of timing at 9000-10000rpm. Is the engine running with fixed timing or is there an advance in your system? Also the spark plug might need to go up or down in heat range once its running better.
You need to put guides or a tensioner on your chain - it is whipping to much - which will be putting variable load on the engine.
The Maha lp13 has a Telma retarder. The Axle holds the weight of the retarder and the load cell measures how much rotational force the Telma retarder is holding back. Let me know if you need some pictures of the setup.
With the load cell you have now leaving it in its current position is probably the best option if a rubber damper is fitted between the load cell and the engine cradle, making a longer arm will increase the amount of error reading as a smaller part of the range is used.
With a longer arm a load cell with a smaller range is needed, like 25 or 50N. But the longer arm could be used as a damper in itself, when it has some weight at the load cell end, the inertia will dampen fast occurring changes in load like those vibrations, just be sure to zero out the static load well.
i can see you are really getting better with your videos! a pleasure for the eyes
You need fuel injection, and with data from sensors you can see theory work. Such as air mass , fuel delivery etc fuel and you can learn this
I like the 10 bit a lot. Gives a good dynamic picture.
Well done Adam. We are all emotionally invested in your success so great to see a win.
Was that an old Amal carburetor that you installed last? Haven't seen one of those since the latr 60's.
OkovPWK, keihin clone.
Awesome! Progress!
I was thinking in the last video that the retarder seems to come in too hard and the PID values might need looking at. Halving the voltage seemed promising and meant that it used more than 1 or 2% of the retarder range. Then you mentioned it doesn't have PID control? Definitely something to look into.
Are you able to hook up an electric motor in place of the engine to get a baseline for the power input and calibrate the software from that? Then you can also switch the motor on from a dead stop and have the retarder ramp it up to a certain rpm over a certain time and hold it at whatever rpm you like, which is what it should do with the engine.
Not sure if you factored it in previously, but the calibration weight hanging from the load cell bolt isn't accurate because of the angle (45 degrees) of the moment arm. The bolt @ 14cm will be much better as it's nearly horizontal with the engine pivot.
Keep up the good work.
Go for a bigger load cell and mount it of the retarder as you stated. Also what is the resolution and sample rate of the software. You need at least x5 the rpm sample rate so at least 1kHz. You could run an accumlator for the pressure for the carb, or run it off the crankcase pressure. The accumulator would just be a bottle and then run that to the carb if you're worried about exhaust gases. Just make the lines short to reduce latency.
Could you use Efi like a speeduino to take care of the fuel delivery problems? Possibly use a smart coil to give an adjustable timing curve.
I think your on the money with the the pid controll not being right less alot less proportional to start with until you can get it to hold the engine at a desired rpm when the load is applied smoothly then start on the intergral so it doesn't overshoot target rpm when hit hard
Thank you for the video! Loving your style of videography and the content is really interesting as well! I look forward to watching your journey
I don't know what percentage you're running but you'll need a bigger float bowl and more than likely needle and seat area. Straight nitro is around 1.7/1 AFR and alcohol is in the 6/1 range so you'll need more reservoir for sure. As far as ignition is concerned, nitro burns slow and will need more timing more than likely.
Try gaping the plug smaller maybe it won't blow the spark out!!!
Methanol burns slowly compared to petrol so will tolerate more advance, and needs to be jetted much richer. I would have thought that 15 degrees is way under advanced and risks running hot. I think this engine has lots more to give and am very impressed with what you are doing. Can't wait to see what you can get out of it.
It might be a good idea to add a lowpass and highpass filter to the loadcell signal. A lowpass filter would probably be the most important
I had to use RC circuit with time constant of around 0,1s to get rid of jumpy signal on my brake dyno.
The chain is flopping around quite a bit. I would guess that it is causing fluctuations in the torque readings. Good luck.
Forget the carb, use electronic fuel injection. Get megasquirt or something like that and an inlet manifold from a fuel-injected motorbike, strip away the extra throttle bodies if you get a 4-cyl or 2-cyl one.
I wonder if the motor was getting to much air from the bigger carb maybe it was both the fuel and air mix at high rpm high load
Loving being on every step of the journey with you 👍
Log format video gives you more leeway in controlling contrast in "badly" lit scenes but that's about it.
Would love to see rpm sweeping using something like an arduino hooked up to a relay to PWM control the retarder to a somewhat stable RPM range (could even make a sweep protocol with that sort of system, especially with a servo hooked to the throttle).
I'm experimenting with log cause almost every shot involves shiny aluminium, hard to expose properly without blown highlights. Not that it's of much significance, but I can't help it. And a complicated process makes it more interesting.
The dyno is hard to calibrate, everyone I know ended up buying a motorcycle dyno. Same with your timing issues, you can put a fuel tech or megasquirt EFI on it and get actual control and Afr.
Load cell calibration from centre of pivot or centre of crank? You may have some weird dynamics going on from remounting the engine in the cradle for the PIP. Also, I think dampening the load cell from that angry little engine is a good idea. Hope this helps. Great content! 👍
i feel like you're probably getting way too much oscillation from the engine. maybe stiff rubber bushings that take up the slop but pad both sides of the load cell? could also put a long bar on the engine to dampen the rotational acceleration (but then you get into wave forms in the bar).
also, under boost do you have enough voltage to spark plug so that it isn't being blown out? could decrease gap to test it.
Use a motor that you figer out the BHP...and then you can compar it ...
Chain slap can give you erratic torque readings