1/8 Rider Ericsson - Home build

[lee webster]

creast, I prefer to use "proper" gates and spues etc. But when I tried to cast a thin section aluminium picture frame using very old salvaged metal meant for die casting, I ended up using the Myfordboy method. It worked and looks good. If there are any bubbles they are either below the surface on invisible to the eye.
Yours is a good casting that will do the job. That works for me.

 

[creast]

Thanks Pat.
Just killed some time and did the lost PLA plaque (x2) in some scrap zinc alloy of unknown analysis.
This time no sagas.
Rich

 

[creast]

I started to mill the crank bracket but my Y axis DRO has been playing up so have ordered a replacement.
The baseplate has now been bored and faced but needs the DRO for the bolt pitch circle drilling.
In the meantime I prepped the walking beam for lost pla casting which required a new slim tube which is too long for my vacuum pot so I resorted to vibrating to help reduce trapped bubbles. This will be the longest casting I have tried using this method.
The piston centre has been part machined and milled for the pivot bosses which I will solder in-situ.
I have decided to try to do the main cylinder as lost pla as it is quite tricky to sand cast due to the bosses at funny angles and to reduce the hassle of supports I have adopted the fabrication method where the bosses and flange are added and glued in place. I will seal the joins with wax before investing.
If this works it will be the largest lost pla I have attempted.

 

[GreenTwin]

Some clever ideas there !

.

 

[creast]

Well, the replacement DRO arrived and has been fitted so milling can continue .
Today I cast the walking beam in a zinc alloy but I wish I had used Brass. This unknown zinc alloy has terrible shrinkage on thick sections and I should have known better really. The part would have been more successful I am sure if I had put the holes in which are on the finished part. This would have given a more uniform section.
However it was encouraging the mould filled being long and slender. I could possible repair the casting but I may re-cast.

The cylinder pattern was prepped and secured into an old food can. The pattern has to be prevented from floating up when the plaster is poured, hence the lolly stick glued to it and the can.
This is currently cooling down from burn-out.
I hasten to add that once the plaster is set it is vital to drill holes in the can base to prevent any pressure build up. First time I forgot, the can pushed out like a pneumatic ram and wedged itself across the kiln walls!

 

[creast]

If at first you don't succeed....
Well as the saying goes it looks like another try again time.
The cylinder casting was a fail but to be honest I wasn't that confident really.
I had my suspicions that vac fill on such a large casting would have its problems, enough so for me to seal the top surface to reduce loss of vacuum but really this needed a proper perforated flask or an extender sprue/riser.
As you can see, the bottom filled great and so did the bosses with no shrinkage faults but the cast failed to fill towards the top.
There are a few bubble blobs but these are not a problem.
Pre-heat of the mould was rather low at 200 deg C but I poured hotter than normal at 760 deg C.
I will repeat but next time will feed right from the bottom and use an extended sprue/riser and possible venting stringers.
Will have to see!

 

[dnalot]

I see in the photos above that you are getting some beads caused by air bubbles in your investment plaster. I struggled with this and have started putting my flask in a pressure pot while the plaster sets up. Squeezes the bubbles down to near nothing. I got the bubbles largely under control by proper mixing of the plaster but sometimes the shape of the part will trap some air. The pressure pot also forced plaster into tiny holes in the pattern or tube shaped patterns. .

Watching this thread with great interest. Keep up the good work.

Mark T

 

[dnalot]

really this needed a proper perforated flask

I am using the solid flask type and have taped several soda straws that travel 2/3s of the way up the flask to aid with the vacuuming of the flask during a pour. Helped a lot.

Mark T

 

[creast]

Mark,
Many thanks for your advice. I will take it on board.
Bubbles can be annoying but better than cavities as they can be dressed off.
My normal investment castings are quite small and predictable in casting.
On this project I guess I am pushing things on a larger scale and things work differently.
I did make a pressure casting rig but this relies on holding the melt in a reservoir under surface tension before pressurising into the cavity.
Any advice is most welcome
Rich

 

[creast]

Not much to report today apart from it's been pretty hot here in the UK and looks to get hotter again.
I re-printed the cylinder and incorporated a large sprue feed straight to the bottom with a spider of gates.
Now the dilemma, do I still use vacuum and incorporate some straws as suggested by Mark or do I vent in a similar way with straws as vents rising from the various pockets to aid flow?
I will ponder on it a while but I would like to see how the addition of straws aids vacuuming.
The base was grit blasted and machined/ drilled etc.
I keep finding more and more errors on the 3D model I have based my work on so since I have progressed this far I will have to make adjustments on the fly.

 

[creast]

Another hot day so I took advantage of the cooler morning and re-cast the walking beam this time in Brass and with undersize hole features to reduce bulk shrinkage problems.
The result was much improved.
There were a few bubble blobs which I expected as I couldn't vacuum degass the mould as it was too long for my pot and of course the dreaded print lines on this particular profile are pretty pronounced.
The original pump on this model is fabricated and soldered but I have chosen to try my cast method and this pattern is made up of five separate parts.
The main cylinder which failed has been re-invested and this time I have inverted the mould orientation and added stringers which when burn out will assist the vacuum fill further up. This will be burnt out tomorrow and probably cast at the same time without vacuum checking for permeability because I am getting impatient!

 

[creast]

Well it's been a while.
The second lost PLA cylinder was a disaster. However, adding the stringers did give me a complete fill!
As you can see though, the casting is rubbish and has serious porosity problems.
My thoughts are twofold, the large sprue would have caused much entrained air and also any hydrogen absorbed in the melt is probably encouraged to come out even more so under vacuum.
I didn't de-gas the melt as I only have the chemical type de-gasser and haven't always seen much improvement in the past.
So, a rethink was in order and after looking at the model in more detail I thought an odd-side moulding of the pattern would maybe work so a wooden pattern was made.
I have never tried odd-side moulding before and the pattern didn't pull cleanly but I decided to still cast it. The surplus areas of sand breakout just means more fettling.
A sodium silicate core was formed in a card tube.
The result isn't fantastic but looks workable. I simplified the pattern to give as uniform section as possible
I did de-gas and flux the melt this time,

 

[GreenTwin]

Nice looking casting.
The shiny surface indicates a good pour temperature.

.

 

[stanstocker]

Some days you dance through a job with great elegance, other days you just slog forward on will power and brute force.

Either way, you kept dancing and pulled it off, so it's a win for the home team!

 

[creast]

Today was spent mainly machining the cylinder casting but no time for photos.
I also cast the pump body, some revised piston links and bearing holder which did not fill due to vacuuming problems during casting.

 

[creast]

Machining the cylinder casting has been a messy and lengthy process resulting in a lot of gritty swarf.
The internal bore details were machined first. Since I didn't create a stepped core to create the water jacket cavity this had to bored out.
Supporting the part on an expanding mandrel allowed the flange and necked section to be turned.
Milling finished off the lugs.
The part will grit blasted prior to the paint job.

 

[creast]

The last couple of days have been spent reviewing how to cast the legs for the Rider.
Originally I intended to lost PLA cast them and had produced the prints a while back. On reflection of my experiences lately on aluminium in plaster casting and the fact I would need a special shaped flask to cast it in I decided to try odd side moulding again.
To do this I decided to print the legs with the apertures filled in and after initial trials added a continuous beam across the feet to act as a runner.
These details will be cut out after casting. The main face is only 2mm thick so quite a challenge I thought on filling issues and the melt freezing too early.
With the Petrobond I found the sand core of the rear cavity was too fragile so I created a Sodium Silicate sand core which would tied into the sand mould with screws which proved be ineffective as the core remained stuck in the pattern but was manually placed back in the drag. The second had sheet metal tabs entrained with wooden lolly sticks glued on to hold more securely.
An extended sprue was used but no riser. The aluminium was cast slightly hot at 750 deg C.
The flask leaked at the joint during the pour which caused some panic but all worked out.
The second casting also leaked at the flask joint but again successful despite excessive flash etc.
Time to make new flasks!

 

[GreenTwin]

I was greatly confused about how to mold windows in casting, but I think I understand it now, and I think I could do it successfully.

As you mention, there may be some concerns about filling thin parts around the windows.

.

 

[creast]

The legs have had their apertures cut out and mount surfaces skimmed. I decided to grit blast these to help reduce the layer lines.
A trial fit to the base was in order and some additional fettling is required.

 

[GreenTwin]

Its looking very good !

.