Blade Expert Impeller

[Müřlıń®]

Howdy guys.

 

I am in the process of training myself to program 5 Axis.

 

I don't have much experience except for the impeller that came with X5.

 

I have found a different impeller that is labled for blade expert.

 

It appears to be more challenging because it has different size blades and one is folded over the others.

 

The model appeared to be scanned surfs that were fairly gapped and I did not see a solid, so I rebuilt the model to a solid to make things better.

 

here's what I came up with.

 

 

I am uploading a file to the FTP if anyone else would like to use it.

 

The file is a zipped folder named BE-impeller, I will put it in the X5_files folder.

 

I had a couple questions to anyone that has had some experience machining these. I plan on using X5 to machine it.

 

Looks like it is about a foot in dia. The blades are very thin. Second it looks like there is not really draft on the side walls. In other words, no liniar surfs to cut with the sides of a tapered mill so one must step it down.....looks like a chattering biach if you cannot imploy thin wall strategies.

 

Where to start...

 

ok when choosing the side walls, would you want to include the sides of the islands as well when picking your machining surfaces?

 

Also would this one not use a tilt curve because of the islands?

 

Since I machined the first impeller with some success, I was planning on just deleting that old impeller geometry and import this new geometry. Then set the clearance zones bigger and make some new tools and holders.

 

Any input would be appreciated....

 

 

Murlin teh rookie.....

[JAMMAN]

I'll be watching this thread for sure, our trunnion comes in next week and though I don't have any 24 blade with splitters and splitter splitters planned I am going to get to play with Blade expert. I hear you don't have to be a 5 axis guru to use blade expert, so I'll fit in nicely.

 

Blade expert itself doesn't seem to draw any excitement on this board.

[robk]

Murlin, maybe you could just send the model out to James and he'll have a machine ready program for you in under 1 hour

[Jayson Kramer]

(ok when choosing the side walls, would you want to include the sides of the islands as well when picking your machining surfaces?) what are you conseding to be ilands in this part. I did notice there are know fillet blends from the vain to the hub were it seems most have at least that for more support?

[Müřlıń®]

Murlin, maybe you could just send the model out to James and he'll have a machine ready program for you in under 1 hour

 

 

Then how would I learn anything :0

 

Perhaps he could chime in and share some of his 5000 hrs of 5-axis programming and beta testing.

 

I think I might have accumulated about 10 hrs with the new morphing 5-axis multisurf paths....

 

Just looking for a couple of pointers....I will machine this model without the aid of Blade Expert.

 

I am sure it is a fine tool and worth every penny but this thread is not about blade expert. It is about learning the 5-axis multisurf functions...

 

Most of you guys have been programming in 5 axis for years...some of us are just starting.

[CJep]

Murlin,

Are you planing to use Blade Expert for this part?

If so it is designed to account for the multiple splitters. If you're working with a solid model I find it helpful to change the color of the various faces in one section that way the geom can be selected by color.

 

The other option is to use the advanced multiaxis toolpaths. What machine are you running this in? Table/Table trunnion rotary, DMU vertical/horizontal?

[Müřlıń®]

(ok when choosing the side walls, would you want to include the sides of the islands as well when picking your machining surfaces?) what are you conseding to be ilands in this part. I did notice there are know fillet blends from the vain to the hub were it seems most have at least that for more support?

 

 

Jay, I took those out in order to build the impeller as a solid. This is not an engineering exercise. I am well aware of the corners needing support. I can add whatever surface fillet I need to, or just let the cutter do the job for me without actually creating the geometry.

 

Ok so each of the 4 blades are machined seperately, then patterened around...thanks...

[slyym]

How do you get to the FTP site? The sticky on top of the page is no longer there.

[Jeremy Herron@DBS Solutions™]

it is is the "Important Topics" that is still pinned to the top of this page.

[Jeremy Herron@DBS Solutions™]

and Murlin, I have this part programmed traditonally too. Let me know when you are done or if you run into stumbling blocks. I will be happy to send you my file, but I would like to see what you come up first, entirely for perspective reasons. Two great minds are better than 1.

 

 

[Müřlıń®]

and Murlin, I have this part programmed traditonally too. Let me know when you are done or if you run into stumbling blocks. I will be happy to send you my file, but I would like to see what you come up first, entirely for perspective reasons. Two great minds are better than 1.

 

 

Thanks Jeremy, I think I have an idea of how to approach it in theory. Now whether you can machine it out of a piece of 7075 without sticking rubber tape on that big blade to keep it from chattering when you finish the other side....dunno bout that

 

There is probably a way to also finish machine both sides at the same time as you profile down to keep it from chattering.

 

Right now though I will pretend that chatter is not going to be an issue ror.....

[Hardmill]

Murlin, maybe you could just send the model out to James and he'll have a machine ready program for you in under 1 hour

 

 

ROR !!!

 

 

Two great minds are better than 1.

 

:headscratch: :blink:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PEACE

[Jeremy Herron@DBS Solutions™]

ROR !!!

 

 

 

 

:headscratch: :blink:

 

 

Sorry should have said 36 minds.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PEACE

[gcode]

Right now though I will pretend that chatter is not going to be an issue

virtual programs are ALWAYS chatter free

[Jeremy Herron@DBS Solutions™]

Thanks Jeremy, I think I have an idea of how to approach it in theory. Now whether you can machine it out of a piece of 7075 without sticking rubber tape on that big blade to keep it from chattering when you finish the other side....dunno bout that

 

There is probably a way to also finish machine both sides at the same time as you profile down to keep it from chattering.

 

Right now though I will pretend that chatter is not going to be an issue ror...

 

 

Morph between to curves on each style of vane... to finish both sides. Create curve on the top edge and offset in xy and up in z for tilt curve....

[Müřlıń®]

Morph between to curves on each style of vane... to finish both sides. Create curve on the top edge and offset in xy and up in z for tilt curve....

 

I knew there was a way

 

I saw a setting for twisted surfs. It let's you use the side of your tool then the draft is not linear. It only comes up though when you use certain tilts.

 

The one also allows for splitter blades....

[Müřlıń®]

I finally got set up last night to run some toolpaths. I started out experimenting with the roughing program. Having problems with splitter blades though. I have been using surfaces to drive should I use the solid?

 

I worked on it for a couple hrs with no success. So I opened the file this morning.

 

Error reading file.

 

The solid I merged for the stock is gone.

 

The solid impeller I merged is now in a format unrecognizable by mastercam.

 

The toolpath box is totally empty...vahished without a trace.

 

The only thing left is all the surfs...bloddy hell!!!

 

Shoulda washed those solids....they came from my laptop and i havent installed MU1 on it yet.

[crazy^millman]

Create curve on the top edge and offset in xy and up in z for tilt curve

 

Murlin look at Opti rough in a 3+2 to rough the impeller blades. Not what Blade expert does, but you might be surprised how well it can do. Thing to think about is the geometry. There needs to be walls extended up from the blades. Luckily you know enough about surface models to do that trick. Then containment boundary for the tool itself.

 

I teach this a little different I tell people to create an offset surface the amount of tilt they think they would like for the side of the tool to stay away from the wall. Then create curve on that edge and there is your tilt curve.

[Jeremy Herron@DBS Solutions™]

I like Ron's approach here. I usually find a good view and 3+2 for roughing. I like optirest if the part has been turned already, then semi the vanes to get what the 3+2 tolpath could not get leaving plenty of stock for then to be rigid enough to cut done later, create offset surfaces of the vanes to protect them while I semi and finish the floor, create a surface of the floor, and trim it from the center of the largest vane to the next, semi and finish the floor, and then the vanes, and deal with the fillets last. Just my way though. I am sure there are others and I would love to here them.

[crazy^millman]

Jeremy, exactly the same way I like to approach it. The ring around the rosy Blade Expert does is suppose to balance the material for cutting, but 1 hour programming time for 12 hours of machining or 12 hours worth of programming time for 1 hours worth of machining? I will be so happy when transform is supported in Machsim.

[Jeremy Herron@DBS Solutions™]

+1000000000000

[Müřlıń®]

Great input guys.

 

Yes I had started last night to split the floor surf up between each blade. I figured that must be the way. I had my chains all layer out to trim to but this morning I had the crashed file.

 

I am unfamiliar with the 3+2 and need to research that.

[crazy^millman]

Murlin, 3+2 is a term used to explain where you put the head if a full knuckle head at at fixed angle and let the X,Y and Z axis move. On a Trunnion machine it is when you lock the 2 rotary axis in a fixed positions and let the X,Y and Z move and on a Head Table machine it is when you lock the Head and the rotary in a fixed position and do the same. To many people want to see 5 axis moves and forget all the kinematics that go into play when doing this type of work. The machine as not as rigid when they are locked into a fixed position and allowed to doing just simple 3 axis work. Problem is getting the tool into an area where it can reach the work you want. So you envision a place where the tool will reach and then make a T/C plane there. When you do then you work there like in a WCS expect the Head is moving to the work, verses moving the part to the place to do the work. Hopefully that was a clear enough explanation.

[Müřlıń®]

Perfect! I had the same thought then...about roughing part of it out in 3 axis

[CJep]

Murlin,

If you're interested I loaded up a multiaxis solution for the floor of the hub. I'm using morph between surfaces using the the two main blades as the morph surfs. The tool axis control is setup to use a tilt strategy with limits active. The main blades are being used in a gouge check strategy tilting the tool away from the blade, the splitters are also used in a gouge check strategy to around them.

 

There is a zip file in the X5_files folder - BE-Impeller_CJep