|396 forum posts|
I'm making a multi cylinder four stroke IC petrol engine, the bores are one inch diameter and the stroke is an inch and an eight, I've no real idea of the RPM, but for this purpose lets say 4000. If anyone can work it out the temperature inside the combustion chambers from this scrap of information, I would be grateful. I've seen various examples online, which range from 100c to 350c. I intend to try and bond the stainless steel valve seats into the Ali head, rather than using the interference fit recommended, as there's seems to be a chemical bonding agent to cover most industrial applications these days, so I would need to select a bonding agent that's recommended to be efficient at temperatures in excess to those in the combustion chambers.
|Mick Dobson||19/01/2014 19:42:07|
|21 forum posts|
I work in the field of testing and developing prototype engines, both petrol and diesel. These are full size ones though, not model scale!
Exhaust temperatures measured at the header pipe, just after the exhaust port, can reach 900 degrees Celcius for a modern fuel injected petrol engine when at full power. I am not sure how this might relate to a very small cylinder as in a model. The fuel is still the same and if using similar compression ratios then the cylinder pressures and therefore bmep will be similar too. The temperature is not constant of course and so the valves and seats will heat and cool rapidly.
I'll do some further digging on this subject.
|Mick Dobson||20/01/2014 17:55:19|
|21 forum posts|
I should explain, the 900C is the gas temperature, not the cylinder head or valve seat materials.
A good estimate of max temperature of the exhaust valve seat is 350C. Also, the seat will normally expand more than the surrounding material so should remain tight. I would go for a light interference fit of approximately 0.001" to 0.002". However if bonding agents can be found that are good to say 450C then this could be worth a try.
|Howard Lewis||21/01/2014 20:39:03|
|3544 forum posts|
Like Mick Dobson, I was involved with I.C. engine manufacture and development for almost all my working life before retiring.
I would be very doubtful if you could find an adhesive/sealant that would withstand the temperatures to which the valves and seats are subjected at full speed, full load, To give you an idea of the temperatures, for many years, valves have used sodium cooled valve stems, and the valve seats are made of materials that cause experienced Methods and Production Engineers heartache to machine.
Despite the exhaust temperature being upwards of 700 C, the valve and seat temperatures will, even instantaneously be well above coolant temperature. And Air Cooled heads tend to run hotter than liquid cooled.
Believe me, if a valve seat comes loose, you will be very likely to find the only repair being to jack up the rocker cover and bolt a new engine under it!
Fit the valve seats by the "Heat and Freeze" technique (In industry the seats are frozen in liquid air or nitrogen before fitting and then pressed in). This assumes that the parent metal is of sufficient thickness to withstand the stresses from the interference.
If possible have the outer edge of the seat inserts overlap the cylinder wall, so that they can't fall into the cylinder. I have seen even the valve inserts (steel in an alloy head) staked, or even screwed in, come loose on air compressors, which run at much lower temperatures, and then they are not a pretty sight.
You are unlikely to run the engine for 50 hours non stop at full load rated speed, but a model aircraft, or powerboat, engine will run long enough to total itself if the seats come loose.
Months to machine and build, but a second to become scrap metal if anything goes badly wrong, so do be careful!
165 forum posts
I'm taking a wild guess that you may be making one of the Hodgson radials? If so have a look at Thomas Bloughs site. He has got round this problem by making the bronze valve seats with a 10 thou step on them and a corresponding undercut in the recess where the seat sits in the head. A hardenened tapered pin is then driven into the seat so that it expands the seat into the recess and is therefore held captive. The angle on the seat is then added afterwards.I have adopted this approach on my 18 cylinder version and have no reason to think why it should fail. I too was very worried about the loctite/interference fit. Tom's site is here.**LINK** and this is the section relating to the valve seats. There is a picture of my progress so far in my album. I can't seem to post one here.
|962 forum posts|
If the seat is deeper than the valve lift it probably will not drop out but just rotate. I know this from experience having had it happen on a single cylinder motorcycle. The seat rotated and since its outside diameter was slightly eccentric to the valve compression was lost. The engine ran well at high revs but............. I must admit on a multi-cylinder engine the damage could have been greater.
On the subject of temperature a model engine is likely to run cooler than a large engine simply because the ratio of surface area of the chamber (through which heat flows) is greater that the volume of the chamber (the source of the heat). However I would not trust any adhesive to hold seats in place. For a model (I am building a Whippet side valve engine) I rather like the convenience of an integral valve seat and guide although it appals me as a retired fluid/thermodynamics engine.
|962 forum posts|
Further thoughts: Any adhesive (which will always be a poor conductor compered with bronze, iron or aluminium) will reduce the conduction of heat away from the seat so leading to a hotter seat. When the maximum temperature of the adhesive is exceeded it will start to degrade (probably burn slowly) and eventually a gap will form between the seat and the head. If the seat does not come loose overheating will increase because the gap effectively insulates the seat from the head.
165 forum posts
Hello Mick, Have you made any progress?
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.