New here & flamelicker ?

[dgoddard]

A tip of the hat and thanks to a few people who gave me some pointers on machining graphite for the pistons. Specifically, that a Surface Feet Per Mminute of 100 seemed to work just right, and multiple who advised about the evils of graphite dust, its mess and its abrasiveness, and an advisory to work it with carbide tools at zero rake angle, i.e. with a scraping rather than a cutting action, and health concerns about breathing the dust. With respect to graphite dust control I have attached a couple of pictures of my Jury-Rigged vacuum cleaner adapter which was powered by my Shop Vac, "bucket vac" (great little vacuum)

This was a quick adaptation and I was not able to do drilling with the tail stock with it in place but that amounted to one small 0.120 hole which did not produce a lot of dust. For the facing, turning, boring, and parting I could witness the dust streaming off the cutter and part straight into the vacuum hose. The way the end of the vacuum hose fit into the wooden support allowed me to push the end fitting as close to the work piece as I cared to.

What little dust did escape was handled by precoating important surfaces with oil and then wiping up the dust trapped in the oil afterwards. Each time a tool was changed, it was sprayed with oil and then blown off with the air hose into the trash can. A final cleaning of the lathe was carried out by spraying the surfaces with oil and then wiping them. Oil is the best cleaner as it will float and suspend abrasive particles so they are wiped away with the oil. A very minimal mess was created although it is probably now time to replace the filter in the vacuum. I washed out the vacuum hoses and there was some contamination but not much.

100 SFPM worked great. The material was very easy to get to precise dimensions and I think this was aided by the rigidity of the material. The cuts came out to the exact amount set on the dial far more precisely than I usually experience with metal in similar setups.

The only problem I ran into occurred when parting the pieces off. When the diameter of the parting cut got down to about a quarter inch, the part simply broke off leaving a slight depression in the surface of the part. I think that this might be partly related to a requirement for the parting tool to be more precisely perpendicular to the part axis than is required for metals, and a parting tool with a wider tip than shank would probably have avoided any problem. However the slight depression where the part broke off was not a major problem to correct but if it were then one should probably part off the work piece with some extra material and then face it off to suit.

 

[dgoddard]

Well, Ridders drawings gave the last detail to the fuel tank / burner, so there is more likely to be innovation. That is the case for me. My first attempt failed because I had difficulty with sweat soldering. My second attempt is shown below. I milled it out of a solid block of aluminum. The central post allowed for a central screw to hold the lid on. The wick tube was epoxied in to retain it. The tank has a chamfer around the top that is likely a little more than decorative. The chamfer allows the tank to snuggle close to the cylinder and place the wick tube directly against it.

As the last picture shows I lack only a wick and fueling it to try it out. However if I am going to provide a video of it, I have a problem that my camera will only produce a ".MOV" file. And I need to know how to convert it to a suitable file, and what that file type is.

 

[dgoddard]

In order to meet the 7mm dimension specified for the top of the wick tube to the bottom of the flame port that was specified by Ridder's, I had to mill a bit off the bottom of my fuel tank and then off the top of the wick tube. A favorable influence on the action of sucking flame in was obvious.

I was able to get a few seconds of operation where the flywheel was no longer coasting. I then tinkered with the working length of the valve push rod by trying several different positions of the adjustable fitting on the rear of the push rod. And I also cleaned the push rod and lubricated it with graphite. The engine was almost running but not quite.

I noticed that after only about a minute the cylinder was getting too hot to touch so I decided to cheat. I used some "circuit freeze" refrigerant of the type used in cooling microchips on a circuit board to find bad integrated circuits. This cooled the cylinder quite a bit. Then engine would run slowly for 2 or 3 minutes until the cylinder became too hot again. While it does spin quite freely (crank and flywheel only coasts for about 2 minutes, and assembled motor for up to 15 seconds) I never got the kind of speed illustrated in several video's I have seen.

Ridder's drawings call for an aluminum cylinder whereas his website calls for stainless steel for the same design. Thermal conductivity of aluminum is much better than stainless so the alcohol flame may be heating the cylinder too quickly where it contacts the front

So I am very open to suggestions from those with more experience.

I did find that adjustment of the pushrod effective length is critical. Apparently The push rod must not be allowed to push the "valve piston" to its innermost position, but only just enough to cover the flame port. Otherwise piston on the connecting rod meets resistance from the partial vacuum in the cylinder before it reaches BDC.

So long as the "valve piston" just barely closes the flame port, if a vacuum develops in the cylinder on the intake stroke, the valve piston is sucked in to prevent slowing of the connecting rod piston. The adjustment appears to be just about right when the front push rod fork is pulled in to within about 0.060 inches of contacting the front of the cylinder, leaving about 0.060 inches of travel under the influence of cylinder vacuum. This is consistent with the description and animation on Ridder's site.

So then, once again please offer your suggestions for further tweaks to improve performance.

 

[ninefinger]

I don't have any experience in running a flamelicker as I'm still in the middle of building mine. What I did notice is the difference in thermal expansion between graphite and aluminum. Aluminum grows about 3 times as much as graphite (12.3x10^-6 in/in deg F versus 4.4x10^-6 for graphite) so on a piston the size of this one the cylinder will grow ~ 0.001" per 100 deg F and the piston will be ~ 1/3 that. So as it gets hotter it will have a sloppier fit which may affect the running (your saying it improves with cooling down lends a bit of strength to this theory)

How hard would it be to fab up a new piston to try - not sure if you could make a tighter fit or try a different material - not sure what else is good to run in an aluminum cylinder...

Just some thoughts..

BTW the whole thing looks excellent - wish it was a present for me then I'd take it as is and tinker with it for days to get it running

Mike

 

[dgoddard]

Mike thanks for the suggestion. I replicated your calculation of the formula
(L)x(delta alpha)x(delta T) and came up with 0.0007 expansion of the cylinder bore over the the piston diameter of 22.0 mm. However the clearance specified in the design is for about .0012 inches. I achieved a very close but free running fit of the graphite pistons.

When the cylinder is hot, if the "valve piston" is pushed in to close the flame port, then a very noticable vacuum develops if the connecting rod piston is pulled toward BDC. Since it passes this test, I do not believe that piston clearance is the problem.

What I suspect is that the thermal conductivity of the aluminum is so high that the flame which is continuously against the front flank of the piston ahead of the flame port is heating the entire cylinder so fast that pretty soon the hot air pulled into the cylinder cannot cool off and contract, thereby defeating the engine.

Ridders made his demonstration model with a stainless cylinder which, among metals, is noted for inferior thermal conductivity. This may be allowing his engine to run longer before suffering the same fate. To run continuously a flame licker must lose heat from the exterior of the cylinder faster than it gains heat from the inhaled flame plus any flame contact with the cylinder.

Measured in BTU per hour per ft per degree F
Thermal conductivity of 6061-T6 aluminum = 90
Thermal conductivity of 304 Stainless Steel = 9.4
So since I have about 10 times as much thermal conductivity I suspect that this is the problem.

If I can block heat transfer directly from the flame to the outside of the cylinder then the superior ability of aluminum to conduct heat to the outside surface and thence to the air should make for a better engine.

Something that at first glance looks like it is working in favor of aluminum is that the specific heat capacity of aluminum is better than stainless steel.

Measured in BTUs per pound per degree F :
The Specific heat of 6061-T6 aluminum = .23
The Specific heat of 304 Stainless Steel = .12

But the problem is that the parts don't way the same, they only have the same size. and you have to use the volumetric specific heat measured in BTUs per cubic Inch To get this, you have to multiply each of the above by their densities and that does not help.

Measured in pounds per cubic inch :
Density of 6061-T6 aluminum = .10
Density of 304 Stainless Steel = .29

So on a volumetric basis for comparing parts of the same physical size but different materials.

Measured in BTUs per cubic inch per degree F
Volumetric Specific heat of 6061=T6 Aluminum = .22
Volumetric Specific heat of 304 Stainless Steel = .35

Which is to say you get 55% more temperature rise for soaking the same amount of heat into the same size part made of aluminum than you would have if the part were made of stainless steel. Since the temperature of the inhaled flame has to be cooled by the aluminum and the aluminum reaches a higher temperature than the stainless steel would for soaking up the the heat of the flame, the aluminum has a harder time getting all the heat out of the flame gases. But with the much higher conductivity it will carry the heat to the cylinder surface faster. The much bigger heat conductivity of the aluminum should overwhelm the volumetric specific heat problem If I can stop the heat from flame contact on the outside of the cylinder from being the main cause of cylinder heating. Then the cooler aluminum cylinder bore should work better. I should just have to keep any heat but what is in the inhaled flame from getting into the cylinder.

I am considering placing an insulating barrier between the flame and the front flank of the cylinder where the flame contacts it so that the parasitic loss of heat from the flame directly into the cylinder will be greatly reduced. The problem is I have not yet come up with a fire resistant insulating material for that location.

I have found some thin exhaust gasket material listed at Advance Auto Parts stores and I will see if I can get some of that to work. If it works it would probably help all flame lickers to run longer before suffering a "hot cylinder"

(NOTE: before I edited this I had interpreted the specific heat advantage backwards. My bad, it is fixed now)

 

[Cogsy]

I wish you luck and I hope you find a solution - especially since I'm trying to build the exact same engine as my first build, also with an aluminium cylinder, and it's also intended as a gift (for my dad)!

Cheers,
Al.

 

[Cogsy]

I did find that adjustment of the pushrod effective length is critical. Apparently The push rod must not be allowed to push the "valve piston" to its innermost position, but only just enough to cover the flame port. Otherwise piston on the connecting rod meets resistance from the partial vacuum in the cylinder before it reaches BDC.

So long as the "valve piston" just barely closes the flame port, if a vacuum develops in the cylinder on the intake stroke, the valve piston is sucked in to prevent slowing of the connecting rod piston.

I just spent a bit of time in my 'thinking room' (read : toilet) pondering this and I have a few ideas.

I was under the impression that the flame hole is to be closed AS the power piston reaches BDC, thereby ensuring maximum intake and negating the chance of a vacuum developing before BDC. Also, I assume that the valve piston should be prevented from travelling any further past the flame hole than the minimum to achieve a seal, otherwise precious vacuum power is lost sucking the valve piston down the bore, rather than sucking the power piston up the bore, then more power is used pushing the valve piston back up the bore further than is necessary.

I have zero experience, and I might be way off base, but this is just what I had worked out for when I get to that part of the build.

Hope something I said helps!

Al.

 

[dgoddard]

I e-mailed Jan Ridders and he confirmed my valve adjustment assessment, but he said that an aluminum cylinder should not be a problem because several have made the design successfully using an aluminum cylinder. His suggestion was that the engine needed cleaning because it was probably contaminated. That did not seem too likely to me until this morning. I repeat my findings which I e-mailed back to him.

----------------
My Response to Jan Ridders Suggestion that Contamination is the Likely Problem
----------------

I was skeptical about cleaning being the basic problem because I could not think of any way in my very limited experimentation with the engine that could have provided any contamination. However when I went down into my shop this morning to make further examinations, I discovered that both the connecting rod piston and valve piston were stuck rigidly. It took only a small force to break them free after which they moved smoothly but perceptively tighter than I had remembered. So there was something in there acting like glue. I then cleaned the bore and pistons using:
1st -- Coleman fuel (Also known as; camp stove fuel, naphtha gas) It is a highly volitile relatively pure component of gasoline. (Definitely do not smoke around this stuff !!! ). It is a superb de-greaser and remover of any forms of petroleum.
2nd -- Acetone

The swab of the cylinder produced a black smudge.

Even more telling was that the head of the piston had a ring of residue on its face all around the edge that had apparently scraped from the cylinder walls by the motion of the piston, with a larger deposit on the connecting rod piston. After cleaning the engine spun very smoothly and freely and coasted well.

This has still left me puzzled as to where this residue has come from. I am using denatured alcohol obtained from the paint store and sold for thinning shellac and as a clean burning fuel for alcohol stoves. Denatured alcohol should be ethanol, C2H5OH.

However there is another potential source of the problem and that is the wick material. Lacking any source of lamp wicks, I tried making my own from what was supposed to be cotton string and then from thread labeled as 100% mercerized cotton. Especially with the "cotton string" wick, I observed that the wick did char a bit more than I thought it should and that there was a crumbly residue in the charred end of the wick. There was much less so in the wick made from mercerized cotton thread. It seems only logical that if there is a polymer component in the wick material that it might melt, vaporize, and condense inside the engine being a sticky fluid while the cylinder is hot and a glue after the engine has completely cooled.

So now I am on a quest to find a proper wick material to test and see if that is the source of the problem. Do you have any material specification for the wick or suggestion of sources for a proper wick material?

----------------
End of My Response to Jan Ridders Suggestion that Contamination is the Likely Problem
----------------

To have put in all that effort to build the engine only to be sabotaged by some residue from the wick, Well ---- that woudl be a wicked thing.

Maybe you should pretend that I did not make such a horrible pun! :hDe:

Anyway, if any of you would care to share with me any ideas of how to obtain proper wick material so I can get this thing running or at least rule that out as the problem, I would be appreciative.

 

[Cogsy]

Because I lack silver soldering capacity, I decided to make my crankshaft differently.

Seeing as you have just constructed this assembly, and I'm about to attempt it, do you think it's possible to machine it from one piece (or maybe 2) rather than make 3 pieces and join them?

I was thinking of machining the O.D. to 34mm, with an 8mm shaft of appropriate length left in the centre, then turning it around and centering in a 4 jaw in the appropriate position of the small spigot and turning that down. Then cut/mill the plate to shape. If the spigot is an issue then I could always make that separate, but I'm keen on making the rest 1 piece.

I'm not sure I have the necessary skills for this procedure, but it would avoid having to locate and solder the pieces. It would be nice to know before I start if there's something I'm missing that makes it impossible or really difficult though.

Thanks for the help.

Al.

 

[Cogsy]

Oops, it appears I asked a stupid question On closer examination of the plans I see the crankshaft is only supposed to be 2 pieces with the spigot soldered on. So I should be able to turn the spigot off-centre and make it all from one piece (if I'm good enough). Fingers crossed.

Al.

 

[dgoddard]

Actually there are a lot of different and very satisfactory approaches. In my own case, since I lacked silver soldering capability I chose to make the crankshaft out of brass rather than steel because, A. I had it on hand, B. I could easily solder brass with lead free plumbing solder. C. The lead free plumbing solder is surprisingly strong.

You may note that I made the pieces so that they interlocked with each other. This made the parts self aligning during soldering so that sophisticated precision clamping was not needed. and any forces would not be transmitted by the strength of the solder but rather through the way the part interlock.

Unlike Ridders design, rather than make a "crank pin" that the bearing would sit on, I made a stub that the connecting rod bearing would be clamped against. The inner race had a diameter of 4mm so my crank pin could not exceed that lest it interfere with some moving part of the bearing. And I found that i had to make a custom screw to have a head that would properly push against the bearing inner race. The screw head is 5m but it has a "washer face" under the head that is only 4mm to properly clamp the bearing to the crank pin.

I got my bearings from, http://www.avidrc.com/product/1/bearings/ and for the connecting rod bearing I used one of their flanged bearings which makes it easier to assure the bearing stays in the connecting rod. All their bearings came greased and with shields. With adequate magnification and a needle you can remove the retaining ring and shield and wash out the grease and re-lubricate with a very light oil.

You will note that I have my bearings held in saddles by bearing caps. Making both bearing supports in a single piece saved a lot of alignment problems because I was able to bore through both sides in a single operation, assuring accurate alignment. However this approach does not provide for keeping the bearings in place parallel to the shaft. I thought the bearing caps would provide clamping for that purpose but discovered that even a very light clamp force on the bearings increased the fricition, which is a no-no for these engines. The solution was to bore the bearing support and caps to be about 0.001" loose and then place a tiny drop of stud and bearing lock in the seats before installing the bearing caps with just snug tightening of the cap screws.

As for options for the crank pin, you might want to consider:
A.
a. Drill and tap a 3mm hole in the crank arm
b. make a 3mm stud that has some unthreaded shank in the middle (might be made from an existing screw or from scratch).
b. run the 3mm stud in as far as the threads allow
c. Install the stud in the crank arm with "permanent" screw sealant.
d. cut or grind the excess stud off on the back side of the crank arm (mabye better done before assembly)
e. make a 4mm spacer to slide over the stud to space the bearing away from the crank arm.
f. use a custom nut or ordinary nut and 4mm spacer to clamp the bearing to the crank arm assembly
This would probably be faster and easier than doing the offset turning.

The procedure that you had suggested produces a "monolithic part", which is interesting, but unless you do something like starting with a casting, you will actually making more chips than part by a ratio of 10 to 1 or more.

Be sure to post progress pictures as well as the final results.

Oh yeah, did I mention I got my graphite for half the price the other places were asking by buying my graphite at http://www.beckergraphite.com/

 

[dgoddard]

Tying to address the wick issue. I ran across the following web page about wick making.
http://www.ehow.com/how_5004463_make-homemade-kerosene-lamp-wicks.html
Once you have found a source of cotton, (that really is cotton, and not some polymer blend that might gum up your engine) you can make your own wicks. They suggest Cotton Balls and Gauze as likely candidates for real cotton. But then they add one trick that I never had heard of. They soak their cotton wicks in a solution of 8 parts water to one part salt. supposedly this prevents charring of the wick. Even charred cotton may provide a contaminating residue. They even suggest making a wick from strips of paper bag, (also soaked in brine).

Of course you could try some supplier of wicks, I found this one:
http://www.oillampman.com/page15.html
But if I cannot find a wick locally I think I will try making one before I order any wick material or pre-cut wicks.

 

[ninefinger]

Would the wicks for an oil lamp be suitable? Not sure of your location. I also have some large wicks from those outdoor patio torches.

A large deparment store with a camping section should have what your looking for.

Mike

 

[dgoddard]

Would the wicks for an oil lamp be suitable? Not sure of your location. I also have some large wicks from those outdoor patio torches.

As a matter of fact the make your own site was referring to making wicks for just such a purpose.

A large deparment store with a camping section should have what your looking for.

Well yes but the nearest store with a reasonable likelihood would be in Springfield MO about 70 miles from here. Or possibly Lake Ozark Area 40 miles from here. Calling ahead is not all that likely to get a knowledgeable response. I suspect Wal-Mart is an unlikely choice as I don't think they stock anything so primative as a kerosene lamp.

But anyway, I found that I have some old "cling gauze" that burns cleanly and appears to be pure cotton, and am making one that way. Stay tuned for an update in a day or so.

 

[ninefinger]

I suspect Wal-Mart is an unlikely choice as I don't think they stock anything so primative as a kerosene lamp.

http://www.walmart.com/ip/FloraSense-Oil-Lamp-Replacement-Wick/16783698

They do have primitive lamps (and the replacement wicks) - but not in camping - it is in the home decor area!

Mike

 

[dgoddard]

They do have primitive lamps (and the replacement wicks) - but not in camping - it is in the home decor area! Mike

Thanks for that. Two issues, the wicks illustrated are flat and I need round, With a name like Flourascent I would be concerned to know if the wick is scented or merely intended so named because they also sell scented oil. If the wick is scented that could be a conaminan, but more importantly in my case I am highly reactive to most perfume scents. However they might also have the round wicks used in smaller lamps and if my gauze based wick does not work out, I will be near a walmart soon and can check it out.

 

[DavidP]

Hi,

Have just found a source of wicks that seems good.
You can buy very cheaply fibreglass wicks for those garden lamps. They are about 1/2" thick.
Simply cut a decent length off (scissors OK), split the mesh holding it all together pull a suitable bundle of fibres and stuff the wick tube. Seems to work well if you don't disturb the original packing of your bundle too much.

I have now run this setup for over 2 hours without charring and without adjusting wick.

Got this idea from the Harris boiler book who recommended asbestos wicks (don't want to go there)

Dave

 

[dgoddard]

....Have just found a source of wicks that seems good.

And that source would be ...... ?????
Is it a source readily accessible to the rest of us.

You can buy very cheaply fibreglass wicks for those garden lamps. They are about 1/2" thick.

So If I need a 1/4 inch wick that shold mean that I can make 1ft of 1/2 inch wick into 4 feet of 1/4 inch wick.

, split the mesh holding it all together pull a suitable bundle of fibres and stuff the wick tube.

It would seem that a loose bundle of fibers would be prone to splay apart a whole lot, or is it twisted or other wise stuck together or do you have some trick to keep the fibers from separating? [/quote]

I have now run this setup for over 2 hours without charring and without adjusting wick.

It is good to know that the incombustable wick material draws the fluid up well. That was my only concern for fiberglass.

Thanks for the input, let us know more about the source of the brand of wick material. An internet search might turn up a supplier with the 1/4 inch size. I believe that the source I noted had fiberglass wick on spools for $10 or thereabout.

 

[DavidP]

G'day DGoddard, (great alliteration)

My source was our hardware chain, Bunnings, which I think is a bit like Walmart? The wicks were in the BBQ section and are designed for those outdoor kero lamps.

Since my post, I have had a look at Jan's site, and note that the length of his wick is much longer than in my application, so in response to the splaying question, quite possibly could be an issue.

I have done a bit of reading about this - am trying to run a 3" vertical multitube boiler from unpressurised spirit. One source (cant remember who so no credits) describes wicks as a valve. The suggestion is that a tightly packed wick will draw less fluid, and provide smaller flame. he recommended 'tuning' the flame size by removing threads from an initially tightly packed wick. That mightreduce the need for a long wick to give a large flame.

My application is preheating a vapourising burner, so not quite as critical?

david

 

[dgoddard]

With respect to wicks. Following the how to make your own instructions on the site noted above.
-- The wick was made with two lengths of "cling" gauze bandage material and opened out to a single thckness
-- The two pieces were layered together
-- The two layers were rolled into a wick that came out a little over 1/4 inch in diameter.
-- A solution of 1 teaspoon of salt in 8 teaspoons of water was mixed in a teacup and warmed it in a microwave to get the salt dissolved.
-- the wick was soaked in the brine until it ws fully saturated
-- The wick was placed it on a saucer in front of a small fan to dry it.
--The wick was stiff at this time and and was rolled it between the palms of the hands -- The rolling and a little tugging at the ends were helpful in reducing the diameter to the size desired and to soften it before cutting a piece to length.

True to the claim made by the website the wick burns the fuel with only a tiny and insignificant bit of charing of any of the cotton fiber.

So as some of our children say these days (especially if they are Harry Potter Fans) "wicked !"
-
-
-
-
-
-
Sorry, I get a pun obsession some times :hDe: