Unusual variations on two-stroke head design.

[Steamchick]

Just one "worry" springs to mind... re: "
I can cut the whole sides of the cylinders out down to the base flange and use a plain 6063 aluminium insert.
This should take the basic ring loading, no combustion pressure. " - What happens when the aluminium rises from cold (20C) to operating temp of the cylinder - maybe 150C? Differential expansion of the aluminium may spring the cylinder apart and even split it?
K2

 

[Owen_N]

Just one "worry" springs to mind... re: "
I can cut the whole sides of the cylinders out down to the base flange and use a plain 6063 aluminium insert.
This should take the basic ring loading, no combustion pressure. " - What happens when the aluminium rises from cold (20C) to operating temp of the cylinder - maybe 150C? Differential expansion of the aluminium may spring the cylinder apart and even split it?
K2

I don't think that will be a problem. I use a jointing compound between the pieces.
The existing steel block I used on the boost side has given no problems.


I suspect the outer cylinder is a fairly high silicon content, but seems more ductile than the inner lining.

That destroyed one of my high speed steel cutters when I was trimming the top of the cylinder.

An hss steel cutter will still nick the surface.

Maybe they have figured out a way to get aluminium carbides in there without nikasil.
It looks like an aluminium finish, but is slightly lighter in colour than the rest of the cylinder.
You can see the contrast at the bottom of the cylinder.
It looks like there are black speckles in it. Maybe hot sputter with carbon particles??
Laser in vacuum?
There is a sharp delineation at the ports.

 

[Steamchick]

Owen, I can only say "suck it and see". Your machining capability looks to be of a very high quality, so I have no doubt parts will fit well. But thermal expansion of the 2 different materials may form a distortion in the cylindrical shape when the motor gets hot. As the engine is so small, it may be in the realms of "less than the tolerance of the engine". By that I mean, how much distortion the engine can tolerate without seizing, wearing badly, or causing blow-by at the rings. Good luck! I am impressed with the head development posts so please continue.
K2.

 

[Owen_N]

Barrel revision R2:

Ports are now staggered slightly, and a cut-down reed block is added,

The drawings are posted here.
there is a need to make the top port roofline steeper than the bottom one.
possibly twisting the ports along their span relative to the barrel may work.
This orientation of the reed valve was chosen because of dimension constraints, and to
clear the bottom transfer duct.
It also gives a straight shot at the top port.
Dimensions are taken from the OEM reed block, but the one selected is from a Yamaha Jog.
I will vary dimensions to suit.
There should be enough clearance for the prop.

There is some revision to the OEM port timing shown.

The actual angles are a bit off due to conrod angularity.
Towards the top of the stroke distance to degrees is more, and towards the bottom, it is less.
The slight lift of 2.5mm in the exhaust should be OK.

There should be little or no drop in power at 6000 rpm, my main load point.
Possibly I should bump the compression ratio up a little.

Non-pipe engines are exhaust open 90 deg atdc +,
Tuned Pipe engines tend to be about 85 degrees atdc.

The transfer blow-down allowance is reduced by 1mm.

The reed valve is to prevent excess blowback of exhaust gas into the crankcase at higher revs.

Reminder of objective:
To provide porting that adapts to rpm and throttle openings.
The top transfer is more ideal for lower rpms and loads.
This should improve engine smoothness at part throttle with a propeller.
This could make a significant difference in vibration with a light airframe.

You probably wouldn't notice much difference with a motorbike- a little less popping, maybe?
road mufflers are good at keeping that quiet currently.

 

[Owen_N]

Owen, I can only say "suck it and see". Your machining capability looks to be of a very high quality, so I have no doubt parts will fit well. But thermal expansion of the 2 different materials may form a distortion in the cylindrical shape when the motor gets hot. As the engine is so small, it may be in the realms of "less than the tolerance of the engine". By that I mean, how much distortion the engine can tolerate without seizing, wearing badly, or causing blow-by at the rings. Good luck! I am impressed with the head development posts so please continue.
K2.

The alterations are below the level of any compression and expansion activity. and any distortion should not affect these activities.
Possibly there may be a little unevenness in piston support during the bottom part of the cycle.

The area between the ports also help with load-bearing during the expansion stroke.

I have reversed the exhaust side for this reason, as there is more wall and piston area on the exhaust side, which is now also the thrust side.
The vast holes cut in pistons nowdays is not good for durability.

Race engines don't have them.
They have disc valves and bridge exhausts overlapping the transfers.

This requires a bigger spacing from the pin to the crown, and long skirts all round.
The transfers are fully outside the barrel as well, and studs are further apart.
You will see this with chainsaw engines as well.

 

[lohring]

Two stroke port layouts have been extensively studied for over 40 years. The pioneer work was done for Yamaha by Gordon Blair at Queens University at Belfast. His #12 cylinder had the worst scavenging while #14 was the best. The final refinement was done by Jan Thiel at Aprilia. Pictures are below. My articles are also attached.

 

[Owen_N]

Two stroke port layouts have been extensively studied for over 40 years. The pioneer work was done for Yamaha by Gordon Blair at Queens University at Belfast. His #12 cylinder had the worst scavenging while #14 was the best. The final refinement was done by Jan Thiel at Aprilia. Pictures are below. My articles are also attached.

View attachment 133342View attachment 133347View attachment 133349View attachment 133348

Thank you for your articles and drawings.

My exercise is not intended as a high power two stroke, but possibly equivalent to a chainsaw motor, but with smoother low power, low speed running mode, as well as reasonable power at 6000-6500 rpm- around 75-90 hp/litre,
with a standard can-type model aircraft muffler and header pipe, about 60mm x 250mm.

I think 4.5 hp at 6000 rpm is about in the range for 60cc.

I don't have a direct standard for this wooden prop , but I think 4.5 hp is a good estimate looking at specs for similar APC props.

If I choose a smaller prop, I can get over 7500 rpm- ie a 20x8 inch instead of a 24x10 inch.

This exhaust does seem to provide a degree of suction, but no tuned pressure pulse.

This kind of running is not a priority with motorcycles or chainsaws, but is significant for model aircraft.

If I could get a 5 hp air-cooled wankel or similar rotary engine, at 1300 to 1800 gram basic weight, I would use that, and this line of inquiry would not have arisen.

This power and weight range in rotaries is not represented in the market at the moment.

Some of the older Sachs units from the 1970's were in that range.
I think one I have seen was about 8 hp.

There have been some made for military drones, but sort of POA, or they have folded.

The military spec setup is somewhat more expensive than a model aircraft spec.

 

[Bentwings]

Owen, I can only say "suck it and see". Your machining capability looks to be of a very high quality, so I have no doubt parts will fit well. But thermal expansion of the 2 different materials may form a distortion in the cylindrical shape when the motor gets hot. As the engine is so small, it may be in the realms of "less than the tolerance of the engine". By that I mean, how much distortion the engine can tolerate without seizing, wearing badly, or causing blow-by at the rings. Good luck! I am impressed with the head development posts so please continue.
K2.

Most of my model aero engines had hemi head or chamber. The old K&B .40 had a bar or deflector on the Ustinov then just a lot in the head . I’m not so sure this didn’t cause detonation. I blew u a number of these while running excess nitro there was a thin gasket that we often took out too that probably didn’t help longevity. All my current gas motors have hemi shape chambers pistons are generally flat top . I don’t have an 4 stroke engines .
Byron

 

[Owen_N]

Most of my model aero engines had hemi head or chamber. The old K&B .40 had a bar or deflector on the Ustinov then just a lot in the head . I’m not so sure this didn’t cause detonation. I blew u a number of these while running excess nitro there was a thin gasket that we often took out too that probably didn’t help longevity. All my current gas motors have hemi shape chambers pistons are generally flat top . I don’t have an 4 stroke engines .
Byron

Actual profile of the combustion chamber seems to be relatively insensitive, if it has about 50% squish, and a bell-shaped cross-section with rounded edges.

A longer chamber likely benefits from twin plugs, for faster burn. I need to trial that, when my twin-spark CDI arrives.
You don't seem to need squish all the way round.

Some chambers work with sharper sides and bevelled ends, in the boost side and exhaust side.
these are machined round to start with.

My next exercise, a skullcap, would likely benefit from a heat-conductive , rigid, jointing compound.
Any ideas? possibly metal powder in a silicate matrix??- waterglass?? Sodium Silicate?
It needs to be stable at 350 degrees C, I think.

The VHT exhaust paint may be heading in the right direction, but you don't want much shrinkage, otherwise it will form voids.
Exhaust sealant looks fairly insulating.

 

[Owen_N]

I am still having trouble finding aluminium at reasonable prices.
Freight is a killer, at about $60 us per piece.
In NZ, no-one is offering to sell small quantities cut-to-size.
I suppose the hobby market is too small, here.

Ex China, I got one piece 6061 dia 60mm x 100mm, $62.84 incl frt and tax (nz) m - about $45 US.
and one piece 7075 40mm x 100mm x 100mm, $104.25 nz, or $74.50 US incl tax (15%)
This is $66.51 usd per kg for the bigger piece, and $56 usd per kg delivered for the smaller piece.
weight-wise?

Considering commodity value is $3.12 usd/kg??
Mind you, alloy, plus retail quantity, will bump this up a lot.
kgs are 0.8 and 1.12.

I think ex shop, cut one-off , 1.5" x 2.5" x 6" was USD $7.20, but USD $60.08 for freight! (ex USA)
total = 67.28 x 1.15 for local tax = $77.32 usd

In kgs this is 38mm x 63 x 150 @ 2800 kg/ cu mt = 3.6 x 10^5 x 2.8 x 10 ^3 = 1.008 kgs approx.
This is about $77 usd per kg delivered (correction)

The China deal seems better, probably due to their cheaper freight rates.

This is delivering mid-march, though, so hardly UPS!
Days roughly 6 + 28 + 6 = 40 days to deliver.
Quite often this is an upper limit, and they can arrive weeks earlier.

-----------------------------------------------------------------------------------
Machining properties:

what are 6061 and 7075 aluminium like to work with?
Do you use this alloy numbering system?

It is common on China sites, but I see the the EU is using something different now.

7075 (alloyed with zinc) seems a lot harder and stronger.
Possibly easier to get a good lathe surface finish?
Does it work-harden?
What is it like for heat resistance?

Some alloys and heat treatments are not good for continued higher temperature use, say 200 degrees C plus?

6061 is usually used for extrusions, and tends to be quite soft.

It is probably difficult to machine nicely on a lathe, and gets bad built-up edge and tearing with HSS tools.
Are tungsten tips any better with this?

I have some tool-grade silver steel and stainless steel here, and it is a bit tricky.
Possibly full flooding with coolant, and tungsten tools, may help.
It hardens very quickly if you drill it with HSS drills.

That stuff can wait until I am better set up with a bigger lathe, a coolant spray system,
and a coolant collection tray.
The coolant probably shoots all over the place, too.

I saw a fully flooded cnc mill setup on you tube, and he had it in a clear acrylic box.

<edit>
my calcs were a bit off:
actual paypal bills were $41.64 for the smaller piece, and $55.43 for the larger piece.
I probably will incur currency charges on top, though.
I will check.
<edit>
back- converting at 1.49,
smaller = 43.40 USD, larger = $54 USD
my initial estimate: 45 and 74.50 was a bit off . maybe I double applied tax at 1.15?
74/1.15 = 64??
so $54/1.12kg = $48 USD/kg, and the smaller piece is 43.40/0.8 = $54 USD per kg.

Still an improvement over $77 usd per kg, total, delivered.

I would have to get someone to collect and re-export it at a lower shipping rate if possible.

Is this very complex re: customs, postage, correct stickers on the package, type of packaging?

Usually this stuff is wrapped in bubble wrap, then put in a self-sealing plastic bag, with a fancy sticker on it.

Then if you want to send it by mail, there is a heavy/bulky goods charge, registered mail charge, etc.
Would it be overall cheaper than $60 USD per kg?

I saw an article on you tube, xjet, where he got a book sent to him by mail, and postage was over $60.
It was a big, heavy book, though.

 

[Owen_N]

Two-Stroke engine skullcap jointing compound:

I have chosen Rocol Steamseal. this is Manganese dioxide, clay, silicates, naptha, linseed oil.

An alumina ceramic or graphite-based compound would be better for thermal conductivity, but it is not easily available online.
This stuff sets harder than graphite-based compounds., and is less likely to drop into the engine.

It is probably too hot an app for epoxy-type adhesives.

I will try running it ,then strip and check for thermal damage, and I can monitor sparkplug barrel temp.

The sparkplugs will be partly threaded into the skullcap.
I will also thread in a few M5 capscrews to help hold it securely in place.
The head really needs more fins, but this setup will do for tests.
I was thinking a 100mm x 100mm x 40mm block to get more finning.
I bought a spare one, but may use it for other things.

Prolonged heat soak of 7075-t7 aluminium plate just drops strength and hardness- possibly getting back to a state similar to T0??

I don't need a lot of strength in 7075, so this seems OK.

I found some comments on embrittlement, but I don't know what they mean by this. Loss of ductility??
Under what conditions?
It seems to be fairly resistant to inter-granular penetration in corrosive conditions.

 

[Owen_N]

More transfer ideas: 1) side-boost port extended from the main transfer, and with 2mm less blowdown and a steep angle. This would blow close in to the rear cylinder wall. The shape gives enough support to the ring with a wider port. It needs a smooth curved transition into the main transfer, and possibly a bit more chamfer. 2) Full reed valve transfers. This is on the theory that exhaust gas blowback and mixing in the crankcase is causing most of the rough running at off- idle and medium-slow running, below 3000 rpm. Reed valves have been shown to be better than piston-port, but worse than rotary disc, in valving efficiency. This should not have a lot of effect on overall power through most of the range, but may cause some residual crankcase charge at the top end. On the plus side, the charge cannot reverse-flow back into the crankcase. It would not work well on a fully piped engine, as pipe suction helps keep the transfers flowing, then supercharges the mixture back into the cylinder on the blocking pulse. 3) Make the transfer blocks more easily removable, with just a thin sealing coat of J-B Weld. This means adding bolting wings, making stud passages, and gluing material between the fins to pick up bolt threads. This does not need to take up much of the fins. This would allow me to trial several layouts more easily. The exhaust timing seems to be quite long already, at 180 degrees. I can maybe lift it 1 mm, increase standard blowdown slightly, and add the offset side boost, with 2mm less blowdown. I am setting the corrected CR at around 9.5, up from 8.0:1 This changes head clearance volume from 5 ml to 4 ml, stroke 36.5, bore 46mm, swept volume 60.7 cc This is good for 3% torque gain, but going from 21mm to20mm for the exhaust opening loses 4.8 % of the effective volume, so there will be no gains from the compression change alone. the exhaust lift could be handy to trial different transfer layouts. For the main transfer, I can stick with 5mm blowdown spacing for now. It will be interesting to see if the anti-blowback for the exhaust gas makes it run more smoothly at the bottom end, and whether more drag in the transfer ports knocks a lot off the 6000 rpm test range. This is where I need to be for this prop I have selected. If I go to a 20x8 prop, I can get a lot more revs, but I don't think it will give me more thrust that the original prop. I have seem 8200 rpm with the 20x8, but the engine seems to overheat at that rpm. I am looking at a 40 x 100 x 100 head design, up from 75x75 at the top fins. A better cooled head seems to keep the barrel much cooler round the back, too. The 30mm tall head I have made doesn't have enough fins. I don't know why the OEMs run the engines so hot, Peaking at 250 degrees C can't be good for them. (measured hot spots). They tend to seize rather easily. Also, many installations are worse for cooling than having the cylinder in open air. Apologies for the drawings looking like they were on the back of a napkin. Actual dimensioned drawings to follow, once my parts arrive. My working drawings still look a bit rough compared with modern CAD, but they do the job.

 

[Owen_N]

Progress update:

The skullcap variation of the head has been overmodified, and now will not start.

I suspect 30 degree plugs with the electrodes only 1.5mm from the piston causes a poor burn on startup attempts.
-A type of flame quench.
I will wait for more head material to arrive, and try a design with a 10:1 CR and 45 degree plugs.

I have lifted the exhaust 1mm, and changed the transfers slightly, which I hope does not also cause starting problems.

I can compare it with my spare barrel/head, which starts well with the current piston and ring.
I may need to convert the spare to swappable heads to be sure, if my next head is also hard starting.

I was also wanting to try low speed running with a compression release restrictor about 5mm up the barrel,
to load up the engine at low speed, which smooths out engine firing.
The hole is there, but temporarily blocked until I get the head sorted.

 

[Owen_N]

Here are the new head drawings.
The head clearance volume should be about 4.4 mls, but these shapes are hard to estimate.

It is sort of egg-shaped in the top view, but fairly triangular in the side view.
How close would I be if I took a segment of a sphere overlaying the side view?
I can extend it out a bit to match the top view more?

The objective here is to reduce the quenching effect of the earlier head, so that it can actually be started.
Do the sparkplugs need to be more out of the squish zone as well?
Usually the single plug is lifted a few more mms.
There have been twin plug heads before, but the head is generally semi-spherical for spark plugs,
and I think the plugs may be slightly further apart, but still about 40 to 45 degrees off vertical.
Glowplug engines are not so touchy.
<edit>
This would require a half sphere with a radius of 13mm.
By slightly squashing the half sphere it would fit.
Minor axis is 18mm, major axis is 28mm.
Anyone know any formulas for elliptical spheroids, to check that the volume is similar?
I will google that.

 

[Steamchick]

I find this fascinating, but beyond my knowledge....
Good luck, but keep trying. That is how breakthroughs happen. I'm sure I am not the only one enjoying your posts, so please don't stop writing.
2 steps forward... One back, is usual in this sort of development.
To check volume, fill cavity with wax or clay, then drop clay into a measuring cylinder of water.... or melt wax into a simple shape You can measure, like a rectangular mould?, then calculate volume.
K2

 

[Owen_N]

I find this fascinating, but beyond my knowledge....
Good luck, but keep trying. That is how breakthroughs happen. I'm sure I am not the only one enjoying your posts, so please don't stop writing.
2 steps forward... One back, is usual in this sort of development.
To check volume, fill cavity with wax or clay, then drop clay into a measuring cylinder of water.... or melt wax into a simple shape You can measure, like a rectangular mould?, then calculate volume.
K2

I just syringe kero in there. It has only one free surface.

I found the formula for a spherical ellipsoid- quite simple..

This looks like an appropriate shape, so I will stick close to this.
"Hopefully" the quenching problem will go away with the different proportions.
with a larger volume to sparkplug size ratio, you can lay the plug over more.
This is what they do in car engines.
Plugs can be about 30 degrees off horizontal.
I have seen single plug two-stroke heads with the plug laid over that far.

 

[Owen_N]

I am back to this engine again.

My block of aluminium has turned up, and I am working on a new head.

I am concerned that this forum may be a little "high quality work" for my bodgie engines tacked together with J-B Weld, though.

What do you think?


I just fixed a bad sparkplug thread (trying to drive a tap in a drill press- overdid it)
I can't be bothered to wait until mid april for proper parts, so I have used a swaged bush, pinned in place with cut-down
1mm nails.
This is slightly marginal. I will see if it stays put.

After the CR check, I will test run to see if it starts OK, then cut the head fins.

I also want to modify the piston and barrel to eliminate the boost ports, using sections cut from old parts.

My calcs show that the piston patch-plugs will stay in place, supported mainly by the skirt hole bridges.
at 6000 rpm, vertical force is about 40N per insert patch, and shear load on the bridges is tiny- about 2 MPa.
support is also via tapering the piston cutouts to resist inwards forces, plus J-B weld.

Most of the piston skirt is cool enough so that the J-B Weld shouldn't deteriorate substantially.

I have used it around the transfer ports, and, apart from slight yellowing, it stands up well.

The objective is to arrest the rapid piston wear a bit, and get the engine durable enough for my next series of mods.

I think the current port mods plus a little higher CR means that no power will be lost without boost ports.

If I carefully fit the piston inserts, they should be at the correct diameter.

The barrel inserts can be levelled with 2 straight edges for support while the initial J-B layer is setting.
The ring passes over J-B bridges OK.

I am assuming that the J_B Weld is around 15 MPa max, but may be a bit lower.
it is flexible and crack-resistant.

18 MPa is high-end for heat cured epoxy.
Most epoxy compounds will be lower than this.

I am working around the 6000-6500 rpm range because it suits the model aircraft propellers.
They would take a more powerful engine to rev faster, and even then, it is more efficient to just increase the prop diameter and pitch.

I wonder if I should go more to small motorbikes, like the 2Stroke Stuffing guy?? (YouTube)
I don't have a great urge to build and fly large rc model planes.

These engines are no good to me unless I can build a contra-rotating prop drive, for a vertical takeoff aircraft.

For "normal" RC planes, I am quite happy with the 200-500 watt range of electric motors.
I would also rather build odd engines and run them rather than go fly any kind of RC model plane, for now.

 

[Bazzer]

"These engines are no good to me unless I can build a contra-rotating prop drive, for a vertical takeoff aircraft."

You do not need a contra rotating prop for a VTOL prop driven model aircraft, just go on the web and see the 1000's of youngsters prop hanging electric RC models .

I would suggest that you stand less chance of VTOL with a contra rotating prop just due to the weight complexities and complications in the system.

 

[Bentwings]

I just syringe kero in there. It has only one free surface.

I found the formula for a spherical ellipsoid- quite simple..

This looks like an appropriate shape, so I will stick close to this.
"Hopefully" the quenching problem will go away with the different proportions.
with a larger volume to sparkplug size ratio, you can lay the plug over more.
This is what they do in car engines.
Plugs can be about 30 degrees off horizontal.
I have seen single plug two-stroke heads with the plug laid over that far.

I’ve done giant scale Rc models for years some very high performance some just for boring holes in the sky.

spark plug holes can generally be repaired with heli coils. Mc master Carr has then many auto parts stores have then unless you are into radically different sizes. I have Tig welded some heads years ago but unless it’s not replaceable it’s not a cheap or easy fix. It maybe that it’s just easier to make a new custom head with your mods built into the design.
Most of my larger gas motors ran about 7k rpm adjusting a little by prop size generally the top is running at a stalled speed so it takes a lot of power to get much rpm change on the ground other than pitch or diameter change. My smaller glow engine ran a lot faster but much smaller props. But the same factor applied. It took a lot of nitro to get much rpm change on the ground. In the air the props began to act more like a screw snd you could get some real speed or pulling power.
mid bevinterested in seeing your counter rotating device. I’ve seen it done but not for a long time. There have been a number of constant speed devices planned but I’ve not seen one that really worked as the real ones do.
byron

 

[Steamchick]

I have used the modern aluminium solder to make aluminium items. It needs care so you don't melt the parent components, but seems as strong as annealed aluminium when made. Very good - I guess - for filling spark plugs holes for re-machining?
Or previously, I would have used zinc alloy repair rods for aluminium.
K2