Welcome! Log In Create A New Profile Recent Messages

Advanced

IC Engine Conversion for Kart Project

Posted by Bill Hinote 
IC Engine Conversion for Kart Project
July 02, 2013 01:19PM
Hi again:

I thought I should reveal the completed conversion of an IC engine for use on my steam-powered go-kart.

After a little research I decided to use this item: [www.harborfreight.com] This is one of a handful of so-called "Honda clones" and quite nicely built.

I disassembled the unit for inspection and mods and discarded items I wouldn't need. These included the obvious exterior pieces but also the entire cylinder head, valvetrain parts including the cam and even the governor parts.

The cylinder features a cast-in-place iron cylinder liner and should be very durable; the piston, pin and rod assembly is very robust and the piston has a substantial crown which will support lift pins reliably.

The attached thumbnail image shows the completed engine in running condition, with temporary lines for running on compressed air. Yes, I retained the original pull starter and flywheel, the engine starts easily and reliably on the first pull and will idle at about 300 rpm on compressed air at 50 psig.

I'll detail the bash valve conversion and the exhaust mods in separate posts.

This project was ridiculously simple to create and the price is certainly right.

Bill



Edited 1 time(s). Last edit at 07/02/2013 01:23PM by Bill Hinote.


bash valve intake details
July 03, 2013 08:15PM
Hello again:

This post will detail the intake portion of the conversion for my go-kart project.

First--when I disassembled the engine and had a first look at the piston, pin and rod assembly I was at first alarmed to find there was insufficient clearance to allow the installation of a single lift pin in the center of the piston because the wrist pin was so close to the underside of the piston crown.

What to do? After some further thought I decided to implement a single-cylinder engine with twin bash valves. The attached thumbnail highlights the robust piston, wrist-pin and conrod assembly with the twin lift pins installed. I'm excited because I think I might be able to demo a high-speed steam conversion as a result of this effort along with other details which will be forthcoming.

Also--my recent conversion of a HF v-twin compressor used a tulip valve; I have become concerned about the necessity of lubricating the valve stem which may complicate things excessively. For this project I have chosen to return to the traditional ball-type valve, but with a twist--I'm using a 1/2-inch diameter silicon nitride ceramic ball which has some pretty amazing specs. Among these is "sphericity" (never heard that term used before!) tolerance of +.00005", -.00004". That's just crazy! I also wanted to simplify the process of assembling the intakes and decided to try and use COTS (that's "commercial off-the-shelf"winking smiley hardware--in this case, plumbing hardware available from McMaster-Carr which is rated for the temps and pressures I'll be using.

The second attached thumbnail shows the valve assembly together, and then taken apart to show the individual pieces. The bottom piece is just a 3/8 NPT short steel nipple which I have chucked up in my mini-lathe and turned a tapered seat into for the ball to seat. The rest of the stuff provides a prechamber and the top cap allows me to put a hardened and adjustable capscrew in place to limit lift. I'm using a lift figure of .035" and limiting further liftoff of the ball to less than .010".

Factoid (thanx Mythbusters!): The silicon nitride ball is 2.5x less dense than an equivalent steel ball so should respond well to higher rpm's. Also, there is no return spring or guide for the ball, I have theories about how it stays centered based upon high-velocity flows around it which force the centering without additional accoutrements. (love that word!)

I believe that 2 of these assemblies feeding a single cylinder will allow extravagant breathing at accelerated rpm's. Testing will certainly reveal whether I'm right, stay tuned.

BTW--for the doubters--this project started and ran from the very first pull on the starter and idles smoothly at about 300 rpm's on 50 psig compressed air. No guides or springs for the ball valves--and the cylinder is inclined at 45 degrees. Ya just gotta believe here!

Enjoy!

Bill



Edited 3 time(s). Last edit at 07/03/2013 08:58PM by Bill Hinote.


exhaust considerations
July 06, 2013 07:12PM
Hello again:

It’s time to discuss what is probably the most controversial part of this engine—the exhaust. Although it’s the simplest part of the conversion to implement (just drill some holes thru the cylinder wall for the exhaust steam to escape out of!) it also affects the performance of the engine in both specific power output and specific steam consumption in surprisingly complex ways.

I took a chance when I bought this engine from Harbor Freight, the exhaust porting was a question mark and I didn’t have much in the way of initial clues as to whether I could create a piston-ported exhaust with it. Fortunately for me there are some nice design details which lend themselves to the desired conversion. Good Luck is mine here!

Specifically, after I removed the cylinder head and all the valvetrain components I found that the as-cast cavity adjacent to the cylinder which handled the pushrod pass-through up to the head and the oil drainback from the head back to the crankcase—was a perfect exhaust manifold—BUT only if certain design compromises were to be implemented to take advantage of the configuration. Stay tuned, read on.

The problem here is, the cavity is only adjacent to about 20 percent of the circumference of the cylinder; the “classic” uniflow engine has drilled ports that are exposed at about 90 percent of the piston stroke and which are arrayed all the way around the cylinder to maximize the exhaust flow. Luckily for me the Stumpf books (available electronically from Google) discuss this issue in depth and a part of that is a proposal to raise the height of the exhaust ports substantially with a number of potential advantages.

My exhaust ports are uncovered at 50% of the stroke!

Here’s a short list of advantages for doing this:

1. The exhaust event is a result of the port area, and the time the ports are revealed as a percentage of a full stroke of the piston/crank assembly. The 50% reveal creates a monstrous increase in exhaust duration and compensates for reductions in port area.
2. The rate of opening of a 90 percent port reveal is very slow and is vanishing to zero as the piston decelerates to bottom dead center; OTOH the port reveal at 50% of stroke is at or near its maximum rate.
3. The action of the piston reversing from bottom dead center and accelerating back upwards creates a pumping action which augments the removal of steam into the exhaust.
4. The issue of recompression of the exhaust is a huge problem with a uniflow, it has advantages in reducing effective clearance volume but narrows the indicator card and thus robs the engine of some of its potential power output; the late covering of the exhaust port by the piston reduces the recompression problem in this application.

Now, the negatives:

1. The effective displacement of the engine is reduced because almost half of the potential expansion cycle available is destroyed by the early opening of the exhaust ports at 50%; of course, this isn’t totally true because the blowdown rate in the cylinder allows for some continuation of the bmep even after the port reveal.
2. The expansion ratio of the steam is cut short, again because of the early reveal of the exhaust ports. This problem is directly related to the as-designed pressures and temps of the steam being supplied and is less of a loss at the lower temps and pressures I’m utilizing.

Here’s the specifics of how I accomplished the exhaust port creation: I drilled through the outer wall of the pushrod cavity, and then further—through the cylinder wall to create the actual ports. The outer drilled holes were then tapped, and threaded setscrews were placed and locked-in with high-temp locktite. The pushrod holes at the bottom of the cavity accepted a tap with no mods and again—setscrews were placed into the holes and locked with partially tapped threads and locktite. The drainback hole was filled with liquid steel putty pushed into to it so it grabs and holds.

The exhaust will be directed upward and through a matching port in the cylinder head plate, you can see the initial holes that have been drilled to match the exhaust cavity. The exhaust will be directed exactly rearward and upward at a 45 degree angle in the engine installation on my go-kart project. The elimination of any further manifolds and piping for the exhaust flow should assist in what I’m hoping is going to be a free-steaming engine.

This engine runs from the first introduction of compressed air (that’s never going to be a problem with a bash-valve uniflow!); the question remains, will it rev freely and produce the power it promises? Stay tuned, it’s gonna be interesting!

Bill



Edited 2 time(s). Last edit at 07/06/2013 11:27PM by Bill Hinote.
Re: exhaust considerations
July 07, 2013 03:10PM
Hi Bill

You could start with less exhaust duration and increase the port height in steps Hard to put the metal back once removed. 50% seams like a lot. The inertia keeps rotation most constant. That means that the piston speed varies and is lowest around TDC and BDC. The important thing is to avoid over compression. You might be right on at 50%. A lot depends on the clearance space.

Andy
Re: exhaust considerations/last details
July 08, 2013 08:43AM
steamerandy Wrote:
-------------------------------------------------------
. The important thing is to avoid over
> compression. You might be right on at 50%. A lot
> depends on the clearance space.

This brings up a last important detail, and I'll button this thread up until I have things mounted and steamed up on the kart installation.

The issue of recompression is very important and should be designed for when setting up an IC engine conversion. In my case the clearance volume was designed into the original IC cylinder head, just slapping a flat plate on to replace it would have resulted in a final compression pressure far above my design operating pressure of 150psig.

Wanting to try and experiment with zero effective clearance volume (that's where the final cylinder pressure is equal to the steam inlet pressure, so no steam flows when the intake valve is opened) I calculated the necessary volume to create the desired pressure; this was assisted by the high location of the exhaust ports as discussed above and reduced the recompression to manageable levels. I was able to create the desired volume by including a .125" spacer plate sandwiched between the cylinder and the replacement head.

With a fixed clearance volume the ideal combination of pressures only operates at one throttle setting (in my case, full throttle); at lower pressures my design parameters begin to be increasingly less optimum. I have chosen to include a prechamber above the ball valves to receive the backflow at lower throttle settings. Additionally, I will be using an auxiliary clearance volume with a controlled port size to the cylinder; the restrictive port connection will allow the aux clearance volume to assist in increasing the effective total volume to 15% of displacement, allowing for smooth running at lower speeds. As the engine rpm increases the port diameter is increasingly restrictive and the aux clearance volume becomes totally ineffective at a speed (which is easy to calculate) that corresponds to sonic "choking" in the connecting port.

Regarding oiling---this engine uses splash oiling. I'm going to "KISS" and just watch the condensation level in the crankcase; because of the "hobby-level" of this project and the fact that I'm not relying on continuous operation, I believe the use of an oil with near-zero emulsification specs and a simple drain-off for excess condensate should be adequate. I'll also be experimenting with a good crankcase breather, it's possible that I can evaporate-off enough of the condensate once things are warmed up, to assist me with this issue.

Bill
Re: exhaust considerations/last details
July 08, 2013 05:16PM
I don't think over compression is much of an issue in this instance. The engine is a classic bash valve, and it is even hoped to run without a valve spring (which may well be possible, I've heard of a couple of instances where the spring failed but the engine kept running). Bump valves are, for all intents and purposes, check valves. Should the compression pressure exceed the steam chest pressure, I would expect the valves to lift due to differential pressure and relieve the excess back to the steam chest. It worked for the Williams brothers, although there would tend to be some loss of efficiency when this happens. That probably can't be helped if the steam chest pressure varies to any large degree unless the compression is set so that the engine never recompresses to admission except at the highest boiler pressure available....which would invalidate some of the benefits of recompression at lower throttle pressures. Anyhow, it's a hobbyist engine and maximized efficiency is one of the tradeoffs for flexibility and ease of construction.

Regards,

Ken
Re: exhaust considerations/last details
July 08, 2013 07:34PM
The problem I was thinking about with over compression is that the compression work can exceed the power stroke work and the engine doesn't output any work. Stumpf noted that the best efficiency is when the compression pressure equals the inlet pressure. With varying power that isn't going to happen except at a specific output. The tuned port aux clearance that Bill is implementing is an attempt to extend the power range. The engine has some interesting features.

Andy
Re: exhaust considerations/last details
July 08, 2013 08:29PM
Thanks for your comments guys!

I was just re-reading Stumpf 2nd edition, the section on compression and clearance volumes and how they're related to so many other things.

There's so much relevant info in there, some of the statements he makes really jump off the page for me. It's fun to be able to relate to this excellent book when viewed from the real-world of my own engine project.

Bill
Re: exhaust considerations/last details
July 10, 2013 02:56PM
Hi Bill

I didn't explain the compression problem right. With high compression to inlet pressure you have vary little range of throttling down inlet pressure before you reach the point were compression pressure reaches inlet pressure on the way up piston position wise as the cutoff point on the way down. And the compression work would be a bit above the down stroke work. You still have 0 to full power throttling range but with a small inlet pressure variation. The aux clearance space should help. Sense you are using a tuned system it will be interesting how that will work. How gradual a change will it be. Will at cause erratic control in that switching RPM range. Hope for some feed back on that. Harry clamed it works great. On the flip side, engine power could be controlled by clearance space. With a check valve allowing over compression to return to steam chest (or line) varying clearance could control output power. Though with a bash valve a check valve may not be necessary.

Andy
Re: exhaust considerations/last details
July 10, 2013 04:38PM
steamerandy Wrote:
-------------------------------------------------------
> Hi Bill
>

> With high compression to inlet pressure you have
> vary little range of throttling down inlet
> pressure before you reach the point were
> compression pressure reaches inlet pressure on the
> way up piston position wise as the cutoff point on
> the way down. And the compression work would be a
> bit above the down stroke work. You still have 0
> to full power throttling range but with a small
> inlet pressure variation. The aux clearance space
> should help. Sense you are using a tuned system it
> will be interesting how that will work. How
> gradual a change will it be. Will at cause erratic
> control in that switching RPM range. Hope for some
> feed back on that. Harry clamed it works great. On
> the flip side, engine power could be controlled by
> clearance space. With a check valve allowing over
> compression to return to steam chest (or line)
> varying clearance could control output power.
> Though with a bash valve a check valve may not be
> necessary.
>
Andy:

You're thinking about some of the right things, but not all of them IMO!

For example, Stumpf says (and this has been proven in fact) that there is an optimum clearance volume/recompression number for each working inlet pressure. You certainly wouldn't run a higher recompression number along with a lower max throttle pressure. In my case for example, the engine idles on 50 psig compressed air and will probably run on superheated steam at a lower psi yet due to the higher energy contained. So I will have an operating range of at least 100psig from full throttle down. Similarly, an engine designed for a max delivery pressure of 300 psig would allow a much higher recompression number and might idle at say, 150 psig--so a 150 psi operating range. Etc, etc. I have chosen to match the recompression value to the full-throttle value of my system based upon the fact that go-karts are most often (?) operated all-in or all-off; for an automotive application it might be more appropriate to optimize for a partial throttle setting corresponding to cruise power at some predetermined speed. Pick ur own flavor here!

Regarding the transition from good flow back and forth to the aux clearance volume, the transition IS smooth, the restriction increases based on flow resistance in an approximately linear fashion until lockup occurs. My Smith conversion of a Mercury outboard motor used this principle to feed the cylinders from the rotary startup valve and you could never even tell it was there.

On my conversion I'll be able to experiment with different transfer tube diameters and even different clearance volumes to "fine-tune" this part of the system.

Bill



Edited 1 time(s). Last edit at 07/10/2013 08:32PM by Bill Hinote.
Re: exhaust considerations/last details
July 11, 2013 01:45AM
Hi Bill

True optimal efficiency is achieved with compression close to inlet pressure. Though less compression would lower efficiency it will give more power. Steam tables such as Keenan and Keyes were not readily available when Stumph wrote his Uno-Flow Engine books. The first printing of Keenan and Keyes was in 1936. There were tables previous to that time but not near as precise or covering the range of pressure and temperatures or all the properties.. Stumpf used simple formula for calculating expansion and compression curves. If you look at the temperature diagram on page 125 of his 1922 book. Note the temperature increases of compression above the live steam and head temperature. That is exactly the kind of thing my calculations show using isenthalpic pressure drop figuring the residual steam state at the start of compression and an isentropic compression to inlet pressure. isenthalpic is normally used in figuring a throttling process. A text book description of an isenthalpic process is two sealed containers connected by way of a valve. One at a higher pressure. Opening the valve results in equalization of the pressures by way of an isenthalpic process. It is also called a free expansion and the engine exhaust process in the Faires Applied thermodynamics book I have says exhaust is a free expansion. At any rate what I am trying to explain is that with a fixed clearance and fixed exhaust close an engine would have a fixed compression and it's pressure curve would be nearly constant. For simplification look at it as a static cycle. The upstroke pressure line follows a compression curve up to the point were pressure is at the external inlet steam pressure and steam is forced into the pre-inlet volume ahead of the engine. So the up stroke pressure line looks vary mush like the down stroke line having constant admittance pressure and isentropic expansion. At some point the down stroke line will match the up stroke line and you would be getting 0 power output. Continued throttling down and the down (power) stroke is totally below the up stroke line and the engine is in breaking mode resisting forward motion. The work produced by the engine is the area between the down stroke line and the up stroke line. When the down stroke line is below the up stroke line the work is negative. Meaning the engine is taking work to turn it over. With dynamics that gets vary complicated Flow in and out of the engine are not constant at constant pressure. They are time dependent and do require some pressure difference to effect the flow. Restrictive throttling does not make for a constant pressure ahead of the inlet. I am working on dynamic steam process simulation to be able to better model a real engine. I have almost completed the steam property plugin for use in VisSim.

Keep up the great work Bill.

Andy


further intake thoughts
July 11, 2013 09:26AM
Hi:

I just wanted to comment a little more about my bash valve assemblies, as follows:

There's considerable doubt in my mind whether something as simple as an assembly of plumbing parts will actually do the job; however, I'll never know unless I try it. So far it works at lower speeds--but the question remains whether an unguided ball will seal consistently at the rpm's I'm hoping to achieve (3000 rpm or more); also, the durability of the mild steel in the nipple I'm employing as the valve seat is certainly a question mark.

I'll run this configuration as a first iteration and if it works I'll be very gratified.

I've considered the possibility of having to upgrade the intake valve assemblies and how this might be accomplished. The issue of the valve seat is pretty easy to solve; Chuck W. has identified a high-quality item available from McMaster-Carr (and elsewhere), it's called a "drill bushing" and is quite hard and should be much more durable if seat hardness is an issue (thanks Chuck!). Also, there are several simple solutions to guiding the ball valve if necessary.

One step at a time.

Bill
HLS
Re: IC Engine Conversion for Kart Project
July 11, 2013 11:53AM
HI Bill
I realy like your project. I might be able to help a little. Use 304 stainless steel for your seat and silicone nitride for the ball. This can be purchased from Mcmaster Carr. The hole for the ball should be 88% of the ball diameter for a perfect seal. Smaller than this and it won't seal well, larger and the ball could stick. A bump valve engine has to operate at very low cuttoff, about 3% and a large clearance volume. The clearance volume can be adjusted using a tuneing tube. This is much better than a large volume over the piston. Let me know the bore, stroke, clearance volume, percent of exaust port opening , operating temperature and pressure and I will give you the tube size for the clearance volume.
It is difficult for a spring to survive A bump valve but it will if it is covered in a tube and there is a pin to stop the ball short so as not to bottom out on the spring. I like a bump valve for the simpilisity but the efficency is better if the valve is timed. We came up with a timed bump valve but it turned out to be more compilacated than a cam lifted valve. This is fun stuff and Bill Ryans go cart was a real runner.
Harry
Re: IC Engine Conversion for Kart Project
July 11, 2013 02:41PM
Bill,

Harry is giving you good hard won advice.

• Lap the ball into the seat with ultra fine lapping compound, then throw that ball away and put in a new
one. Old trick with White pump ball checks.

• Minimal clearance volume and definitely use his clearance volume tube, a very clever idea; but for wide
rpm range it needs its own control valve if you build a much more sophisticated engine later.
Low compression for slow running and high compression for high speed.
Figure out your expected average rpm in pulses per second, then match the resonate frequency of the
tube length to match that, or be some half or quarter length of that frequency. Full length is better.
Closed pipe organ prime frequency determination; but with hot steam there is a temperature/frequency
correction .

Built two bump valve engines out of old outboards with in line ball checks and both ran beautifully.You will be surprised how well such a simple valve can be and still work well.
This ought to be fun, keep us posted

Jim
Re: IC Engine Conversion for Kart Project
July 11, 2013 03:54PM
Great progress Bill - you don't waste time! Very informative and interesting discussion too.
Mike
Re: IC Engine Conversion for Kart Project
July 11, 2013 03:56PM
Mike Clark Wrote:
-------------------------------------------------------
> Great progress Bill - you don't waste time! Very
> informative and interesting discussion too.
> Mike

Thanks Mike. That's because I'm in a hurry, as I get older the press of time becomes more urgent.

B.
Re: IC Engine Conversion for Kart Project
July 12, 2013 02:09PM
Here is a plot of the speed of sound in steam over and isentropic expansion from 150 PSI,600F to 14.696:


thanks
July 12, 2013 02:54PM
Hi guys:

Thanks for your inputs, much appreciated.

My first goal is to get things running. After that we can address "refinements".

I suspect I'll have my hands full (and more) just chasing leaks and getting boiler controls refined; there may be major or minor changes in hardware or design required too.

Refinements come later.

Thanks again,

Bill
Re: IC Engine Conversion for Kart Project
February 14, 2014 01:22PM
(I'm new here and this is my 1st post, hello everyone!)

It seems like an obvious idea to me, but since I haven't seen it mentioned, I thought I'd ask anyway:
You're mentioning the recompression of the exhaust as being a problem, however since your stream engine was a 4-stroke IC engine in its previous life, did you consider keeping the exhaust valve mecanism and using it to solve that problem ?
Of course the camshaft drive would have to be changed to run at the same speed than the crankshaft in this case.
In fact the full exhaust could happen by the valve; of course we'd get further from the uniflow design than you may have originally wanted, but the 4-stroke IC to steam conversion would potentially be easier...
Re: IC Engine Conversion for Kart Project
February 14, 2014 03:55PM
Vincent Wrote:
-------------------------------------------------------
> (I'm new here and this is my 1st post, hello
> everyone!)

Welcome and thanks for your interest in my project.

> You're mentioning the recompression of the exhaust
> as being a problem, however since your stream
> engine was a 4-stroke IC engine in its previous
> life, did you consider keeping the exhaust valve
> mecanism and using it to solve that problem ?
> Of course the camshaft drive would have to be
> changed to run at the same speed than the
> crankshaft in this case.
> In fact the full exhaust could happen by the
> valve; of course we'd get further from the uniflow
> design than you may have originally wanted, but
> the 4-stroke IC to steam conversion would
> potentially be easier...

There have been several conversions of 4-stroke engines to run on steam; most of the time the cam is modified so there are 2 lobes for each valve to compensate for the cam which runs at 1/2 crank speed.

Also, there have been a number of schemes utilizing auxiliary exhaust valves to relieve excessive recompression in uniflows, and your suggestion is one of the possible solutions. J. Stumpf's 2nd edition of "The Una-Flow Steam Engine" (dated 1921 and available free through Google electronic books) covers this subject extensively.

However, there is another line of thought which has many supporters (myself included): The recompression portion of the cycle can be beneficial if properly designed and utilized. It creates an effective zero clearance volume--and clearance volume is considered a loss of efficiency. The volume is adjusted so the recompression value approximates the operating pressure of the system. In my design I have established the clearance to create a recompression equal to full throttle--and then at part-throttle settings the recompression value is actually higher than the supply pressure, creating a backflow scenario through simple impulse ("bash"winking smiley valves which are essentially one-way check valves except when activated by the lift pins in the piston near top dead center.

The indicator card that results shows reduced power output for each revolution of the engine due to some of the output power being absorbed by the recompression process. It is generally accepted that this type of engine makes up for the loss by operating at higher rpm's. In fact, the bash valve engine when properly done has many characteristics similar to an IC engine. I'm hoping to duplicate or exceed the rpm range of the original version of this engine.

Hope this helps.

Bill



Edited 1 time(s). Last edit at 02/14/2014 09:00PM by Bill Hinote.
Re: IC Engine Conversion for Kart Project
February 14, 2014 06:42PM
Thanks for the insight, I pretty much expected it was way too obvious to have been overlooked.
I just got a pdf of the book and will do my homework; thank you :-)
Re: IC Engine Conversion for Kart Project
February 14, 2014 08:07PM
Vincent Wrote:
-------------------------------------------------------
> Thanks for the insight, I pretty much expected it
> was way too obvious to have been overlooked.
> I just got a pdf of the book and will do my
> homework; thank you :-)

No problem, man!

We're all learning something here--and part of my goal in posting so much information is to show that there's no such thing as "failure". Instead, we learn from our mistakes and they become the foundation of better results.

BTW please do refer to my thread entitled, "Bash valve concept musings" and which details the focus of my current effort. The failure of my first attempt at intake bash valves has diverted me but ultimately should result in a better running motor.

I hope.

BH
Sorry, only registered users may post in this forum.

Click here to login

All files from this thread

File Name File Size   Posted by Date  
MSE 4 done 4.JPG 256.1 KB open | download Bill Hinote 07/02/2013 Read message
piston and rod with lift pins.JPG 284 KB open | download Bill Hinote 07/03/2013 Read message
MSE 4 valve assembly.jpg 259.7 KB open | download Bill Hinote 07/03/2013 Read message
Stumpf 125.JPG 77.2 KB open | download steamerandy 07/11/2013 Read message
SOS_600F.JPG 61.2 KB open | download steamerandy 07/12/2013 Read message