FFFF metal parts

From: Jon Hunwick (Delcam  International PLC)
To: Gregory Pettengill (Cote Art & Engineering)
Date: Wednesday, July 27, 1994
Forwarded to RP-ML by Greg, 94 08 03
Subject: FFFF metal parts
Dear Greg,
Thanks for your mail.

As you can see from my main message to the group, CARP has several partners. 
At least two of these (DELCAM and WEBSTERS) have extensive experience in the 
mould and toolmaking industry. An early trial in the project involved just 
such a process as you describe. Without going into too much detail (as the 
part is confidential) it was an oil gallery for a four cylinder petrol 
(gasoline) engine. The part had draft surfaces all over the place, and the 
aim was to make a corebox to make the sand cores needed to cast the holes. 
As the part included a long vertical drop (with tapers) it was decided to 
make it in 6 bits. These were the top and bottom of the box (cope and drag) 
and 4 loose pieces which slotted into the cope. The CAD model was broken 
into the necessary parts, and all the extra surfaces added to form the box. 
Each part was then made on FFFF (by the LOM process) and the resultant 
'wood' patterns finished off by traditional pattern-making methods. 

Although the project was a success, it was found to be only marginally more 
cost-effective than traditional NC machining as the part was so complex. (We 
define complex as having doubly curved surfaces - rather than complicated 
which we define as being feature rich). We concluded (and proved later to 
our satisfaction) that a more cost effective solution is to build the part 
in FFFF and then take resin castings from that. 

All of this is fine, of course, unless you have a part which would (in 
plastic injection moulding) need side cores (slides, up-and-aways, lifters) 
etc. etc. to actually get the tool to open. Building a tool this way, even 
with FFFF, is a lengthy (and costly) business. In such cases casting 
**direct** from the FFFF model, even if it is destructive, is a very 
attractive option.

So - what I'm eventually getting round to is YES! I agree with you, and we 
have both considered and tried the approach you suggest. 
Provided the part is largely prismatic there is no great difficulty.
As soon as you have undercuts or overhanging features you run into problems.

Keep in touch,
Jon Hunwick