What causes belly?

[glbreil]

I am noticing that my parts have some belly in them. I am cutting 8.25" tall 17-4 and the hole that I cut will accept .700" gauge all the way through which is what I want , but on each end a .701 gauge will go in about .1/2" to 3/4" before stopping.

I am cutting submerged using .30mm Cobra Cut A, a rough cut and 2 skims.

Any ideas what might cuase this or what I might change to get rid of it. I is not a real problem, but I would rather not have it just the same.

Thanks Gary

[WILLEO6709]

using beer for flushing????? seriously...checked your wire squaring block lately?

[Dan Fox]

8.25 inches and only 2 skims? On my AC100's, I use 3 skims, even on the short stuff I cut - usually 20 - 50mm.

[glbreil]

The hole is square and the same on both ends so I don't see how the squaring block could be an issue.

Dan my finish is plenty good with 2 skims and I guess I have thought of more skims, but its not like its to small in the middle, the ends are to large already and I figured more skims would just open up the middle if anything.


What do you think?

I should probably leave well enough alone, these things get bored about 1.5" on each end anyway.

Gary

[precisionmetal]

I believe it's an electromagnetic thing that happens, and shows its head as the guides get a long ways apart.

On my Agie, I have a setting I can tweak to adjust this though to be honest I haven't messed with it. I'll usually take 3 skims like Dan does if I want a tall part really straight.

[Dan Fox]

Gary-

A third skim wouldn't make your opening larger; instead, it would go in between your rough and first skim. You'd have to adjust the offsets, of course, and make some test cuts, but you'd want to end up with the same finished size, just straighter.

But you're right. Not worth messing with on those parts!

Dan

[wiretechjack]

The belly in your part is caused by flushing. When you are cutting taller parts you are going to get a belly in the rough cut. This is due to the lack on flushing in the middle of the block. On Charmilles machines you can adjust the AJ to try to keep the belly to a minimum.

The amount of belly that you are talking about isn't that bad in the thickness you are cutting. You will have to decide what you can live with and how much time you want to take to get rid of it. More skims will help if you have to hold tenths in that thick of a block.

Talk to the tech support people for you machine and they should be able to help you with settings that will reduce the belly in the rough cut.

Jack

[glbreil]

Thanks Jack, that is probably it, however Agie has long ago lost interest in consulting about my machine - Agiecut 100. I guess that makes it even more interesting.

Thanks for the help! Gary

[kazoo2]

Gary, belly is not caused by electromagnetics, flushing (although this can be one component of the cause), or the kind of beer you drink (or use as dielectric). Belly or bow in a tall part is caused by wire lag. Wire lag is inherent in the WEDM process, and is a result of a combination of wire tension, cut speed, wire feed rate (how fast the wire comes off the spool), flushing pressure, part thickness, etc. As the wire moves thru the cut, these physical forces tend to make the wire bow, thus causing the error you are seeing as either belly or bow in the work piece. You can reduce or eliminate this by slowing down the cut speed, speeding up the wire feed rate, and/or applying more tension to the wire. You can get more detail about this at these links -

http://www.makino.com/about/webinars...Profitability_

http://www.makino.com/about/webinars...ot_in_Wire_EDM

These are links to a couple of webinars on Makino.com that address this issue specifically. The first link (webinar) has a simple pictorial that shows wire lag and explains what causes it, and how to reduce or eliminate it.

More passes WILL reduce the belly or bow, but that will also increase your cycle time.

Jeff

[Jay Cee]

I would agree that belly can be cause by wire lag on a rough cut only or a rough cut with a surface finishing trim type situation.
However, when using multiple skims, I was taught as precisionmetal suggested.
The charged wire carries an electromagnetic field that allows it to be pushed away from the workpiece.
It is most evident in the center because it is the furthest distance away from the tensioned pivot points (the guides).
This explanation explains why you can end up with bow in the opposite direction.
This explanation also explains why an electrical modification (AJ on Charmilles as wiretechchat explains) can eliminate this error.
I have used this AJ adjustment on my Charmilles many times and know it to not to be a mechanical adjustment (wire tension or wire speed).
It is purely electrical...

Adding another pass will help, but you must be careful what kind of pass you are adding.
It must be what I call a "geometry pass".
Another "surface finish" pass will only shine up the error and make it look prettier.
Your "geometry pass" will look more like rougher settings with a set feed-rate controlling the speed rather than a wide open servo governed by resistance .
It will also be looking to bite off a couple of thousandths depending on your wire diameter.
Tall pieces like what you describe would require at least two of these "geometry passes" if looking to eliminate bow.
These passes will also ignore the error caused by your rougher's wire lag thus giving you cleaner geometry everywhere (sharper corners and directional changes).
As mentioned before, "Surface finishing" passes will only make the error look more shiny.

[Brian Pfluger]

To All,

Everyone has a bit of truth in their explanations, but the ultimate fix or correction varies depending on the flavor or brand of BEER (machine) being used. Belly is generated MOST in Roughing Operations, and is MOSTLY (maybe 80% plus) generated by mechanical forces. It's obvious to see that the least accurate section of any WEDM'ed part is at the center mid-point of the parts thickness. This is where the wire is farthest away from the guiding system, and where the combating forces of Wire Tension & Flushing take their toll on part accuracy as we "PUMP" our Beer Kegs for greater speed!

Once the Belly error is introduced into the part, it can be difficult to remove...especially as work piece thickness increases. Any error (belly) that remains in the part past the 2nd Skim Cut will not be able to be removed by adding more passes to the process. The final skim cuts only improve surface finish and do not have enough power to remove this error, and simply explained...will only follow the existing vertical shape as a result of capillary action.

As far as how to go about reducing/eliminating Belly from parts, this will change per machine model. The methods that have been explained (AJ or Servo Adjust values) all have to do with the Spark Gap distance...meaning that you are changing the effective distance the discharge sparks have to jump in order to erode the work piece material. This also changes the Spark Gap voltage and machining speed slightly, but think of changing these values like this:

Die Cavity Example

Belly Outward (Barrel Shape - Big in Middle Section): is the result of the Skim Cut speed being too slow...meaning that there was too much material (LMC) removed from the part. To correct this, the machining speed can be increased, or the AJ/SV-Adj value can be changed (lower the value for Makino machines), which does the same thing as slightly increasing the machining speed.

Belly Inward (Hour-Glass Shape - Small in Middle Section): is the results of the Skim Cut speed being too fast... meaning that there was too little material (MMC) removed from the part. To correct this, the machining speed can be decreased, or the AJ/SV-Adj value can be changed (increase the value for Makino machines), which does the same this as slightly decreasing the machining speed.

-Brian

[Jay Cee]

Brian,

I'm not saying, I'm just saying...
Why would so many of us been taught to make the Voltage adjustment instead of the servo adjustment if they are in effect doing the same thing?

The other 20% you mention... Is that a result of a magnetic field?

Jay

[Brian Pfluger]

Jay Cee,

The Servo Adjust or AJ settings do control the Spark Gap Size and resulting Voltage, and I also only play with these settings on the 1st Skim Cut and NOT the Rough Cut. I am sure there are some major differences in the settings between the different machines, generators, and machine vintages. Over the past 10-15 years, much has changed in generator design and function for all EDM OEM's.

As for the magnetic field thing...I've heard very little of this before, but I would be interested in seeing this data and research. I know it is possible to "PUSH" the wire away from the work piece during the Rough Cut if you have too high a power without removing enough material with your program offset. This is like an aggressive No-Core/Coreless process, wherein you create a BIG TIME Bell Mouthing condition in your part. I have been able to overcome these issues by changing the flushing, wire tension, and offset settings in the Rough Cut. Any final straightness tweaking (+/-0.0002" range) is done by adjusting the Servo Adjust (SvAdj) settings in primarily the 1st Skim Cut...at least on the Makino’s.


-Brian

[Jay Cee]

Please correct me if I am wrong here as I am only a guy who makes these machines move.
I, by no means, am an electrical engineer or even someone who's done copious amounts of research on the subject...

My understanding is the servo adjustment is basically a speed of linear movement adjustment.
It would effect the spark gap only the same as a dwell would.
In the example of a slower servo rate, more material is being removed because the charged wire is left close enough to spark for a longer period of time.
The Aj (voltage) adjustment is an actual electrical adjustment where as the spark is allowed to travel farther do to increased voltage.

Is this correct?