What's new
What's new

Burn time for tall part, many noob questions

BridgeportinD2

Hot Rolled
Joined
Feb 28, 2012
Location
Iowa, USA
Fellas-

Being a total noob to EDM, I've got some questions for you old hands.

My foreman brought me a two-piece job to wire yesterday, and this is definitely the tallest job I've done in my short time on the wire.

It's a simple job, cutting a 1/8" wide slit through the side of a flanged cylinder relative to it's center.

The cross section of where I'm burning is: the flange height and thickness(where it terminates at the cylinder wall) is 3/4" x 3/4". The remaining cylinder wall measures 8-1/4" height x 7/8" thickness. That works out to being approximately 31-1/8 square inches full-slot for the in and out passes.

Note- No super fussy tolerance or surface finish, as this is just a crush slot. Just rough burn.

So, I'm needing reassurance/guidance on the burn rate on the tall detail of this part and my resistivity readings.

I watched the screen when the cut went from the flange to the tall portion, and noted it went from running at .061"/minute to .0048"/minute, which didn't really alarm me. I changed the power settings a bit to get a better rate. Wound up burning at .0052"/minute comfortably.

Noted resitivity at the start of the burn, it was 64,000 ohms/cm, and at times during the day I checked the rate and resistivity. Rate stayed right at .005/minute, resistivity was at last check 540,000 ohms/cm.:eek:

That resitivity reading alarms me. Almost 8-1/2 times from when the burn started.

Machine- Sodick AQ327L, company bought new in 2006.

Wire- Sodick Hightech .010" brass.

Material being wired is 4140@42 Rc, if that matters.

Changed filters, water, and resin tank 1 week ago.

Cleaned both upper and lower conductivity pieces and indexed them to a new surface at the same time.

What are you fellas thinking on this? Am I a worry-wart that needs to calm the hell down or do things look out of line?

Many thanks guys.
 
Been a while since physics, but resistivity is usually expressed in ohm meters, or ohm centimeters, and in this case is a measurement of the resistance to current flow of your dielectric. More resistivity means LESS conductivity . . . so your DI resin tank is doing it's job. Low resistivity is bad.
 
Hi Bridgeport:
This is a topic of great and enthusiastic confusion in the wire EDM world.
Here's a recent thread that shows what I mean:
http://www.practicalmachinist.com/v...tivity-sky-high-after-cutting-inconel-335126/

The OP posted, as you have, that his water RESISTIVITY had gone up when in fact, his water CONDUCTIVITY had gone up and the solution in his case was to replace the DI resin to strip out the excess ions that had accumulated in the water.

It doesn't help that the EDM machine manufacturers sometimes label the meter stupidly; on my machine it calls the measuring device a "conductivity meter" but the units are those of resistivity (ohm*cm) and the displayed value goes down as the DI bottle stops working, so until you think it through, it is pretty misleading.

Red James has it exactly right; high numbers of conductive ions reduce the resistivity, and the job of the DI bottle is to bring the resistivity back up by pulling the excess conductive ions out of the water and binding them into the resin.
You can poison a bottle relatively easily, and although it's expensive, it's no big deal, even though you may have to throw away not only the DI resin but also the water if the poison persists even with a new bottle.
A common culprit is skanky coolant residue or oil on your parts from previous machining operations.

I wouldn't sweat it at all; replace the resin and see what the water does, but first be aware of what your meter is ACTUALLY measuring.
If your resistivity is TRULY high, you have a different problem and need to look elsewhere other than replacing the DI resin.
The problem will usually be in the conductivity probe or its associated electronics, which then triggers the machine to run water through the DI bottle when it's not supposed to.
Short the legs of the conductivity probe with a screwdriver; if the value shoots up, the probe and meter are measuring CONDUCTIVITY.
If it shoots DOWN, the probe and meter are measuring RESISTIVITY.
If nothing happens, the probe or the electronics are hooped.
Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com
 
Noted resitivity at the start of the burn, it was 64,000 ohms/cm, and at times during the day I checked the rate and resistivity. Rate stayed right at .005/minute, resistivity was at last check 540,000 ohms/cm.:eek:

At least on my Sodick AD325L, the meter reads resistivity (for those wondering if he is seeing resistivity or conductivity). I've gotten it up past what the meter will even read by scrubbing the tank, changing all the filters and DI resin at the same time. There is a bit of a delay between trying to change the value and the value changing on the screen.

You can always call sodick to verify, but if the resistance is getting too low I add DI water, and if it's high I add plain table salt. You probably had that 540,000 value when you started cutting which is why the cut is slower (.005/min is pretty slow for a rough cut). What were your conditions? When you get back to the machine, add a couple packets of table salt to your work tank and drain and fill it a few times to cycle some water through. If you haven't cleaned the conductivity sensor, you might want to do that too. Wait a while and see if your value goes down. Shoot for somewhere in the 70,000 ballpark.

EDIT: I'll PM you my cellphone number. If you have any questions about it, just send me a text. No phone calls at work allowed though. Otherwise Sodick is very helpful at their normal contact line which is (847) 310-9000.
 
Red and Marcus, thanks for the replies.

Marcus-

Ever thought of writing a book on EDM? Your posts are always easy to read and packed with excellent real-world experience. Always worth reading.

As far as the resistivity probe and it's operation, I'm pretty sure I know it is functioning as it should, and that it does measure resistivity, and not conductivity.

A few months ago I started a thread about issues with our Sodick, and one of the fellas suggested cleaning and checking the probe while I was cleaning the work and supply tanks.

I cleaned the probe, but did not know how to check it, so I just started running the machine (pressure from above to get it burning).

The toolmaker that trained me on the machine made an observation one day while I was burning a part that'd been run by us several times. He thought it was taking a long time to burn, like nearly twice as long as in the past.

He admits to not being very attentive to the details of the machine, and was further confused by the control and readings on the screen after we had Sodick come in and update software and do some tune-up work.

Unfortunately for he and I, no information on the changes made to the control were given. He didn't know whether the resitivity reading was good or not, as the values had changed.

At the time he noticed the long burn time, he figured we'd better change the resin. The reading was at 5,000 ohms/cm before the change, never moved while we were running. Within a few minutes of running the machine after the change, the count started climbing and leveled out at 64,000 ohms/cm. Burn time was significantly reduced. Apparently that resin was used up completely.

Running jobs with different materials or work piece sizes, I see the reading fluctuate by +- 8,000 ohms/cm while burning, but it always seems to return to around the 60,000 mark, which is lower than when the resin was new.

Thanks for the link to the thread about the Inconel.

Any thoughts on the burn rate?

Thank you for all the input. You're helping me to learn and not be all by myself in the dark in this.
 
gking86-

Thanks for the advise.

So it seems the 64,000 or so reading is in-line with what you've experienced. That's reassuring.

I don't recall what my exact conditions are, but will let you know tomorrow. You're wanting the on/off, MAO, SV I assume. Any of the others? Haven't messed with other than what I listed. Maybe I should.

Have found that the on/off settings given when programming are way too aggressive, and I always seem to have to drop the on and up the off to keep the wire from burning in two. How is it with your machine?

Thanks again for helping me out.
 
Nothing of value to add to this thread, except (the "tall" part reference) (;)) I had to burn a part 15" tall once. Well, once meaning one part, cut into 8 pieces. It sucked.:drool5:
 
Hi BridgeportinD2:
Thank you for that very nice compliment; I really appreciate it.
Actually though, I'm a noob myself; I'd never even seen a wire EDM run until 2008 when I just couldn't resist anymore and bought one just because I HAD to have one.
I did a lot of head scratching in the beginning too, as we all do when we're learning the ropes on something new, and I still have tons to learn, so I hardly qualify as an expert.

On a related note, there are a lot of guys who are really worth listening to on Practical Machinist, and I've received a lot of super advice for which I'm eternally thankful.
The time and frustration it saves is immensely valuable to me, and it's nice to be able to give something useful back from time to time.

With regard to your burn rate, I agree with gking 86 that 0.005" per minute seems slow even on an 8" tall cut, but so much depends on all the variables for each individual cut that you may have hit the upper limit that's defined by the part geometry and your ability to flush the slot.
Sometimes you're better off to accept a slower, stable cut than to endlessly jigger the parameters and suffer a gazillion wire breaks.
How many inches you're going to cut really defines how much tuning is worthwhile, so I've trained myself to push for best possible performance only when I judge it's worth it overall.

I usually plan my wire work as "set and forget" and my business does quite well that way, even though I could gain time in the cut by fiddling more with it.
But I'm a busy guy and it's sometimes more useful to me overall, to be able to walk away from it and not have to babysit it.

So trust your gut; when you think it's burning well and is stable; maybe best to just let it run.
Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com
 
Hi BridgeportinD2:
Thank you for that very nice compliment; I really appreciate it.
Actually though, I'm a noob myself; I'd never even seen a wire EDM run until 2008 when I just couldn't resist anymore and bought one just because I HAD to have one.
I did a lot of head scratching in the beginning too, as we all do when we're learning the ropes on something new, and I still have tons to learn, so I hardly qualify as an expert.

On a related note, there are a lot of guys who are really worth listening to on Practical Machinist, and I've received a lot of super advice for which I'm eternally thankful.
The time and frustration it saves is immensely valuable to me, and it's nice to be able to give something useful back from time to time.

With regard to your burn rate, I agree with gking 86 that 0.005" per minute seems slow even on an 8" tall cut, but so much depends on all the variables for each individual cut that you may have hit the upper limit that's defined by the part geometry and your ability to flush the slot.
Sometimes you're better off to accept a slower, stable cut than to endlessly jigger the parameters and suffer a gazillion wire breaks.
How many inches you're going to cut really defines how much tuning is worthwhile, so I've trained myself to push for best possible performance only when I judge it's worth it overall.

I usually plan my wire work as "set and forget" and my business does quite well that way, even though I could gain time in the cut by fiddling more with it.
But I'm a busy guy and it's sometimes more useful to me overall, to be able to walk away from it and not have to babysit it.

So trust your gut; when you think it's burning well and is stable; maybe best to just let it run.
Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com

OP, if you haven't heard this yet, wire edm depends on 3 things - flushing, flushing, flushing. :D Seriously though, bad flushing is the number one 'suspect' on a slow burn time IMO. I used to burn keys in bushing that would often start terrible, then do semi-ok once they were in the cut, but never really 'good' (non-submersible mind you) and my boss could never understand it. :crazy:

I am not familiar with your machine, so this may not apply. On the older Charmilles you could get a kit that allowed auxiliary flushing, it was basically a 3/4" loc-line that screwed onto the upper head that had an independent water supply (from the upper and lower heads) that you could adjust for flow/direction/nozzle type and point it inside a round hole/bore, etc. Maybe check into that for the future, if you are not using a submersible that is...
 
First I would like to agree that Marcus should start an "EDM for Dummies" book. (LOL) Yes your posts are understandable and your "screw driver test" was a perfect way to test that!

Secondly, I agree completely in regards to flushing! If you can get the sludge out and away from the cut you don't have to fight it and the actual cutting of the metal which will increase speeds given settings etc. are good for your machine conditions.

That being said, I have been accused of being anal or paying too much attention to trivial details in regards to the way the burn looks on the part as well as the slugs. I look at the sludge and if it lies evenly or thickness of it. The rainbowing on the part around the cut. To me it can show if flushing is good/bad, if the power setting is too high, water conditions. Does anyone agree with signs your slug and part can tell you?

I found it true with milling as well. From the way your chips look, pushing of metal look on the part, or even rusting on parts. Those things can help you better adjust your speed and feed as well as sharpness of tool and even if your coolant flow is adequate or has too much water in mixture are just to name a few.

Has anyone found that to be true on EDM or even payed attention to it?

Thanks,
Toni V
 
I always look at the slugs to check the flush and HAZ. Anybody else check the slug for bow or belly ?
 
Hi JTB3 and tvalenzuela:
I look too, but only to confirm why a burn didn't go well if I've had problems.

The trouble is, if I diagnose poor flushing from the pattern of cut debris on the slug for example, there is really nothing I can do about it if the flushing was already running wide open, except maybe open the servo gap and that seems to have very little effect on my machine.
Similarly, if the slug is not straight, I can do some things to improve it, but on a roughing cut I will skim anyway, and I rarely do one pass burns (but that's because of the kind of work I do).
I know you can tweak burn parameters to mitigate barrelled or hourglassed slugs but I don't typically find it to be worth it for my kinds of work because production burning is rare for me.

I'm reading between the lines here, but it sounds to me as though you are getting really worthwhile gains from diagnosing the slug and tweaking your burn settings based on what you see on the slug.
I'd love to know a bit more about what you both are finding and how you correct for the things you do find.

With some jobs I'm already content just to get through the job without a gazillion wire breaks, and the killer for me is things like cross holes, and geometric section changes and impossible distances from flush cup to job surface, so there's not much I can do about those.
In those circumstances I tune my cutting conditions to try to reduce the frequency of wire breaks and accept that I cannot work to set parameters and expect a known outcome.

I find I rarely run to the numbers and expect my job to come out correct; again almost always flushing related.
I do test squares on some jobs, I do in-process gaging on others.
Often I find I need to uses successive skims to bring a part feature to size and I can judge pretty well by now if a skim cut is taking a tenth or if it's taking two tenths.
So I treat the machine almost like a precision grinder capable of fine cut increments on complicated contours; much more like a woodworker fits his parts together than how a metalworker measures and cuts to a specific number.

Since I prototype so much, and since my customers often have no idea what they really need for fits and clearances, we're often fiddling our dimensions and tolerances as we work, so they get good numbers they can put on the production prints.

So please, if you will, let us know what you look for on a slug, and what you do when you find it.
I'd love to learn even if I don't end up using the new knowledge all that often!!

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com
 
I find myself checking slugs more on taller work peices . I'm mostly checking to see that my flush pressure is fairly even. If not I can adjust the upper and lower indapenditly. I also check the bow or belly of the slug. This allows me to adjust the final trims accordingly. I have these tecs all saved back so not much test cutting needs done even on fussy work.
I know we have sent slugs out for recast testing before as well.
I also find myself keeping slugs to use for set ups or making spacers.
 
I find myself checking slugs more on taller work peices . I'm mostly checking to see that my flush pressure is fairly even. If not I can adjust the upper and lower indapenditly. I also check the bow or belly of the slug. This allows me to adjust the final trims accordingly. I have these tecs all saved back so not much test cutting needs done even on fussy work.
I know we have sent slugs out for recast testing before as well.
I also find myself keeping slugs to use for set ups or making spacers.

Big slugs make smaller parts ...

Small slugs make heel-blocks ...
 








 
Back
Top