1920 Redrup Radial - replica motorcycle engine

Me too. Super cool! Good to see you got the corebox demolding issues sorted out!

Jeff

 

Thinned drywall compound seems to work well for many lost foam practitioners (foam losers? )...

But now that you have your corebox basically dialed in, it's hard to imagine why you'd suddenly change methods..

Jeff

 

A scratch build engine project......I like those. Just a couple quick comments.

1. I'd suggest you use 1.3% instead of 2% for aluminum shrink factor. If you use 2% your part will be oversize by approximately .007in/in. On a small part like this it's probably insignificant but when you cast larger parts you'd be off~ .070" on a 10" dimension.....too much for me!
2. It's certainly true that the higher the melt temp the more susceptible to H2 porosity, but I think it's much harder to control furnace tune with waste oil burners and you want the most lean/oxidizing burn possible and it that's harder with waste oil than say propane. However, with you will expose your melt to a very high volume of superheated steam that comes a long for the ride in combustion air. Given your Southern hemisphere, I'm guessing it's Summertime and humid. It's why my furnaces are resistive electric. Many of my YT subscribers that follow my lost foam casting videos experience porosity and think foam is the culprit and it never is and almost always a result of inadequate melt management.

I (have been) subbed!

Best,
Kelly

 

Shrink can vary modestly, but it correlates strongly with melt temp. Once a metal freezes the shrink goes essentially as the coefficient of linear thermal expansion (or contraction in this case ). If you are ever in doubt about shrinkage for a given metal, look up the coefficient and multiple by the difference between melt and ambient temperature. This will give you a dimensionless factor that can be applied as in/in, mm/mm etc. Of course, the CoLTE also varies subtly with alloy. Alloying reduces melt temp so pure aluminum shrinks more than its alloys.....same for other metals. This is why pour temp does not affect shrink factor because shrinkage only begins at the point of solidification. That's not to say that pour temp does not affect shrink defects.

I have not because I do not presently own a 3D printer. See the thread rocco linked.

My small furnace was 8kw and I recently upped that to 10kw. My large furnace is 22kw. Both furnaces can operate at 1/2 or full power. I use 1/2 power for heat treating after furnace is at temp. There are links to my furnace builds in my signature. Yes, 3kw can be enough but you will find the (first) melt time is a strong function of your furnace mass, thus my efforts to make very low mass furnaces as my casting sessions are frequently single melts. My smaller low mass furnace will melt 10lbs of Aluminum in an A10 30 minutes (now closer to 20 minutes with 10kw) from a cold start. I think this is about as good as it gets with resistive electric unless you devise a way to incorporate heating elements in a pure refractory wool design, then slightly better. By comparison, when it was a high mass furnace, it was 3x that, but the second heat was .7x.

Just a guess for 3kw-A5, but if you did a very good job of managing (low) furnace mass, probably 30-60 minutes to melt.

From watching your vids, I realize you want to remain as faithful as possible to the original engine design, but you'll have your work cut out for you casting the cylinder/block in iron. Have you calculated the casting weight? Then add 20-40% for a feed system? -Much easier in aluminum with sleeved cylinders from your starting point.

Makes sense for preheating for same reason as discussed above regarding furnace mass.

It's not as easy as it may appear.....even harder on a scale as small as a crucible. A tubular lance that just bubbles Ar won't be very effective and may even be counterproductive. You need a very fine dispersion of the gas. Hobbyists try all kinds of things for degassing Aluminum and they always conclude that they're being effective when dross appears on the surface, but simply agitating the melt in the presence of air/O2 creates surface dross. The more agitation the more dross. The arguement always goes "look at all the dross that brought up", and my reply is always "look at all the dross you created". As a control, degass with a tube and Ar, and then do exactly the same with air, and compare results.

Degassing Lance | The Home Foundry

Best,
Kelly

 

Martin is a sharp guy and very experienced and knowledgeable foundryman. If you read my link you'll see clear signs I borrowed many aspects from Martin's design.......my method of construction and choice of materials was different, and an improvement IMHO. I'd also point out that Martin uses that lance in a rather large dip-out crucible furnace, not an A5!

An A20 weighs 12lbs. It will hold ~60lbs of iron. Plus your shank weight. With the blind heat intensity of iron, that will require lifting equipment or at minimum a two-man shank with large separation distance and/or shielding.

Good luck. My advice would be to put that to the test before you commit to a direction. If you can find someone willing to pour it, all the power to you. If not, decide if you're willing build the equipment and the learning curve for iron for one casting.....this is the voice of experience speaking to you.

Though that is well insulated and relatively low mass for a dense refractory furnace it is not a low mass furnace (not even close) by my standard. You don't get to ignore the mass of the base and the lid. It must be heated too. In fact, the furnace refractory I recently replaced in my 22kw low mass update had a 5/8" dense refractory wall. You need to read my links and the links in them!!

My small resistive electric furnace has been Insulating Fire Brick, Dense Refractory, and Moldable Ceramic Fiber. IFB is a reasonable compromise of mass, durability, and ease of construction. I used K23.

I have a dedicated 200amp/240vac (48kva) service in my shop. If I wanted more, the utility would install it and even heavily subsidize the cost of doing so, so they could sell me more electricity, as long as it's single phase. I have 3ph high voltage 300 yards from me, but they won't make that available to a residence, and even if they would, the cost of running the lines would not be subsidized, be very expensive, and the cost of 3ph power is much higher than residential 1ph. I pay ~10 cents US per kw/hr. So for 30 minutes @ 10kw it costs me <5 cents to melt 10lbs of aluminum. My large furnace is similar per pound cost. I have a rotary phase converter and have no complaints at all. In fact, It allows be to buy great second hand industrial machinery for pennies on the dollar because besides being outdated, nobody wants 3ph equipment, and it means I pay residential rate instead of business/industrial.

Funny, because same thing goes for natural gas. When I built the large furnace it was originally intended to be NG fueled. At the same time, I was building a patio and wanted NG fire pit and other appliances as part of the project. At first I was denied those features on my building permit because in their eyes, my service was not adequate. They arrived at this decision by surveying all NG appliances on the property summing their maximum demand and comparing to my present NG service capacity. When I pointed out this would never be the case they didn't care and told me to call my utility......so I did. Turns out, the limiting factor was the size of my meter and regulator. They told me as long as my neighborhood infrastructure was adequate (which it was) I could have as much NG as I wanted. So I added a 500kbtu/hr NG burner to the list and they upgraded my meter and regulator for free. Residential NG pressure is only a few "H2O in the US but the other side of the regulator and meter is 50-100psi. I asked to tap that, and the answer to that was a flat no for residential zoning........I tried .

Best,
Kelly

 

Melt practice, feed system, (and casting design) the cause and cure for all ills irrespective of casting process.

It would be worth your while to invest more time in understanding the AFS (unpressurized) and Naturally Pressurized feed systems. If you are a real Campbell disciple, like Bob Puhakka, the latter is the preach. There is a very long thread here on BiFilm theory with Bob's commentary.

Here's some random observations and comments:

1. You could make things much easier on yourself using more draft and generous fillets on things that can afford them, like the pouring cup and riser extensions. Why saddle yourself with what's typically considered the minimum draft?
2. Why such a small pouring cup? The only downside to a larger one is more metal usage. If you remelt your returns do you care? Versus the penalty of not being able to maintain a continuous pour and fails?
3. I don't like round pouring cups or round sprues. -Neither does Campbell. The worst pouring cup and sprue combo is a round funnel and round sprue because they aspirate and entrain air due to Coriales affect. I initially used round sprues and cans but with lost foam, there is absolutely no reason for a round sprue and it's easier to make a square tapered sprue. If you are not bound by processes and tooling that incorporate them, why use them? My initial pouring cups were roundish tear drops but all since are square cornered to the extent possible. They are all offset with weir. I was amazed at the improvement it made in my castings. Those defects half way up your riser could be H2 shrink defects or could be entrained air from turbulence in your feed system.
4. I think a deeper flask would be helpful. Instead of worrying about exothermic toppings, why not just make them blind vented risers at the proper height so they are more insulated and without the fuss of making and adding extensions? You also have little head pressure at the top of your mold cavity (bottom of the piston skirt) because it's so close to the height of the top of your pouring cup. That skirt is the thinnest part of your casting and the last to get (the coldest) metal. You might also consider some scratch vents on the core to generously vent the mold there. Many of your attempts are incompletely filled through the piston skirt. You attribute this to pour temp which I'm sure is a/the major factor but no doubt you could pour cooler with more head and good venting.
   
5. Why are the risers positioned where they are and not at the more massive wrist pin section of the casting? .....and why is it connected to the entire height of the piston? Looks to me like the thin connecting web is likely to freeze before they can do their job which to me, is a question mark where they are located. I sort of doubt whether they're doing much for you. I'd position them adjacent to the wrist pin bosses, move them a little closer if possible and/or connect them with a thicker web, from the bottom (top of piston) and only to the height of the wrist pin boss. This way it will feed the bosses and head of the piston which are the clearly most massive part of the casting. Since these areas are the first and last to get (the hottest) metal. it will also promote directional solidification from the skirt down to the head which is exactly what you want.
6. I realize you're trying to reduce the mass of the wrist pin with that small core but in my opinion it's a waste of time given its diminutive size, and furthermore, can just be a source of core outgassing in one of the most susceptible areas of the casting. In fact, the entire part could be made with no core as match plate cope and drag. You'd probably need some guide pins for extraction. Looks like you may have deleted the small core on this go-round. Are you (still) baking the large core and keeping them warm and dry until the pour?
7. Maintaining green sand and molding with cold sand is problematic. Green strength generally suffers as the sand becomes cold. A sharp corner in a green sand mold has no strength and becomes a source of debris in your mold. When I did conventional sand casting, I filleted everything including the gates and runners. It's not always possible on all features but it is always helpful in reducing sand inclusions. You can include traps in your feed system but that does nothing for sand wash originating from within the mold cavity. I don't have to contend with any of this in lost foam casting.....just keep my sand dry. Not sure if your shop/molding area is heated, but if not, you could mull your sand, bring it inside for the night with a lid on the bucket, and mold just before the pour.
8. I may have missed it but are you still using waste oil throughout the melt or did you switch to Propane?

Not trying to be a critic, just helpful

Best,
Kelly

 

Can't imagine why. It's so small it does very little to reduce the mass of the section and like I said, just another potential source of problems.

In addition to the above, eliminate the undercut like 99% of pistons and no core. Like Bonz said may need a gagger to pull reliably without fallout, especially if your sand has low green strength.

Could still bring a bucket of molding sand inside for the night. Alternatively, I think Fishbonz made a heater for his sand. IIRC, it was either part of an electric skillet or maybe one of those heating elements for lighting charcoal he could dial back. He may also have put one of those reptile aquarium heating pads in the bottom of insulated cooler. Bonz?

......Or live with it and take what you get.

Best,
Kelly

 

I get wanting to make a faithful replica, but you only have a couple drawings and photos so how faithful can it be? Design nuance aside, you are using CAD, 3D printing patterns, using modern alloys, will no doubt use modern machining equipment, .......etc, etc. It's certainly not 1920 methodology and there will be many other engine parts that will vastly benefit from modern materials and processes that weren't available in 1920. Also, I'd hazard a guess to say you can't be certain of the quality of results that were actually realized in 1920. The aluminum alloys of the era were poor by comparison to today. Sure, in theory, the undercut in the pin boss saves a few grams of rotating mass, but in practice it's in an engine that is almost certainly unable to materially benefit from it.

Almost anything is possible with right combinations of skill, time, and money........doesn't mean it's practical. Things should be made a simple as possible.....just not simpler. Obviously, your project your choices. If you don't make some practical compromises, you may not be able to complete the project. Much greater challenges are coming!

This is a small relatively simple part. If the goal is learning, maybe one lesson is introducing that much SS core of uncontrolled cure to a small molten metal mass, that you want to have high integrity, may not be an optimal strategy. Ask Martin why he prefers shell cores.

I think the best orientation is skirt up and feed the most massive portion of the part first.........with the primary benefits being it allows easy access to feed the preferred locations and naturally promotes the preferred directional solidification. The majority of the mass will be there irrespective of orientation so why not use it as opposed to fight it?

-Carry on

Best,
K