On Machine Inspection w/probe

[behindpropeller]

I'm curious if anybody does on machine inspection with their probe?

Any hints, tips, etc?

Tim

Similar Threads:

• Anyone using a probe to perform on machine inspection?
• Used Machine Inspection Questions
• Tool Inspection Machine
• Used machine inspection

[Delw]

anything +-.005 or higher I wouldnt have a problem with.
remember your machine made the part so if its off then its going to be off even though it checks good.
flatness roundness parrellism I wouldnt even think about checking on a mach that I ran the same part one.
for example, lets say you ran a circle of 4" and you need to be square and round, obviously your machine will check it perfect for squareness and roundness cause it made the part to your program that was square and it was a perfect circle in your program.
even if you have a place that has a .005 flat spot in it its going to check good cause the flat spots was in your ball screw. or your tables was out of wack etc etc.
NOTHING and I MEAN NOTHING Replaces a INSPECTION ROOM.

Delw

[Caue]

Some customers here use this feature successfully, even for inspection or modeling (to copy some part damaged and stuff).
It's also possible to take out the results by DPRNT command.
You can set your process to measure a sample according to the time (clock) or the quantity of cycles (parts counter) and so on...

If you have some specific need, please let us know!!

[behindpropeller]

Here is a thought.....

We have lots of parts that are +-.005. If certain features could be checked with the probe it could save time. Obviously if the machine is wrong, the probe check could be wrong.

I was thinking that it might make sense to put a known standard on the table bolted down. Have the probe check that standard every time it starts. If the standard numbers are not correct, then have it alarm out and not inspect the parts.

TIm

[Donkey Hotey]

Your probe could probably still do a fine job of inspecting. The ruby is a different diameter than most of your cutters so, a bad spot on a ball screw is highly unlikely to be in the same spot as a feature. You're also going to probe features like bores in a cross fashion, not circular, so the backlash conditions will be similar but, different.

The probe can also find conditions like undercut from thin walls deflecting under load, dull cutters, etc. I'd say that you could do a fair job of inspecting a lot of features in the machine. It's obviously not a CMM but if your parts required that, you'd already have one.

My question would be: if you already have critical parts in the machine, and you're thinking of using a reference standard, why not do in-process probing of critical features, then adjust and do your finish passes in the program?

[Delw]

I guess I look at it this way, would you take a hammer and nails and use your inspection grade granit plate as a work bench?

using the probe for reverse engineering parts is completely different then using it for inspection purposes.
I guess the bottom line is if you ask your customer if they would accept parts checked on a machnie with the probe and they said yes, I wouldnt see why not.
Maybe I am old school but I just dont see how this would be the correct way of inspecting a part.
I think it would be better maybe if you ran them on one machine and checked them on another? but if thats the case you might as well buy a cmm for much less.

Another way to look at it is would haas use there probes on machines that are running parts to do inspection on there parts for the machines eliminating there inspection Dept?
if this can be answered 100% yes then maybe you can trust it.


Delw

[behindpropeller]

Guys-

I'm not a job shop or machinist. I'm a mechanical designer that has the opportunity to use our TM-2 for prototypes....etc. I'm looking at some light production with on machine inspection.

My idea is to start manufacturing our simple parts that we are paying out the nose for right now.

I can make jigs to check hole locations but would like to pull some numbers and spit them out as part of the inspection.

Any money saved will potentially go into tooling.

Tim

[Delw]

Something like that would probaly be ok sence your the designer anyhow, and you know what your looking for.

But you brought up a very good subject and I am glad you did that. I would love to hear the pros and cons about it from bigger shops, areospace shops, commercial shops and inspectors as well.

the reason I am glad you brought this up was I have heard haas salesmen and used machinery salesmen say you will save alot on inspection equipment by buying a haas with a probe, they insinutate you have a very accurate cmm built in the machine( which you do in a sence). this bothers me. Face it anyshop that is setting up KNOWS there is a huge cost in inspection equipment just for the basics alone, alot of machinist working at shops then thinking Hey if I buy a machine with a probe I dont need to spend money on costly inspection equipment and cane easly start up my business for alot less. this will get more cheap hacks into machining that are clueless about everything else involved. They buy a used machine for 30-50k with a probe on it stick it in there garage and undercutting machine shops like china does.
Oh yeah dont get me wrong I am not knocking Garage shops as when I started my place in the early 90's I was running 5 omniturns and hardinge AHC's with 6foot barfeeds out of a 20x20 garage while working a full time 12 hour a day machinist job.

When we bought out haas I was pretty much told the same thing, I been looking into buying a CMM at the same time. That thought immediatly jumped into my head thinking Damn I could get this haas and not get a cmm makes it more justifiable. and I have over 25 years of machining experiance in the areospace feild. inspection grinding lead screws thread guages pretty much all phases etc etc.

that wasnt a key point in the decision of this haas but what I am getting at is I actually thought about it when we were talking about buying a new machine, we all try to find things to speed up production.( kinda like a pair of calipers vs a dial bore or micrometer)
After running the machine and playing around with the probe a little( the machine makes perfect parts) I realized that it might not be a good idea to use it as a cmm, and I looked at it this way. if I had a probe set up on my fadal would I use it to inspect parts hell no, dont get me wrong the fadal makes great parts as well, but because of its ages there are certain things and ways you have to program the parts to make them run accurately. most fadal owners know that running a circle on a fadal with a endmill is not accurate (unless you have glass scales)not even close to what the same year and houred haas will do on circles.


Delw

[Matt@RFR]

Del, no offense intended here, but how beat up are your machines that you won't trust them to anything less than +/-.005"??

Personally, in my home shop with my '07 VF-2ss, the probe and machine are more accurate than most measuring tools that I own. I mean, the probe is calibrated to a gauge ring (+/-.00005" or so?), and when is the last time your machine didn't repeat to +/-.0002" in position?

Now, I'm not saying that you're wrong, but if a small shop doesn't have a dedicated metrology lab, I'd think you could go a long way with these probes as long as they are calibrated correctly, the machine is in good condition, and you use a little common sense. I would dearly love to have a CMM, but I don't, so...

Edit: You and I posted at the same time Del. I'm not saying that the probe will replace a CMM. Not even close, but I really feel that it will be good to a reliable +/-.0005". I don't quote work that requires tolerances like that because I don't have the resources to verify the probe. And, I for one buy measuring tools whenever I can justify it, and I realise that the probes are not an end-all piece of equipment, but I do think they are much more accurate than you are stating.

This is an interesting subject. Hopefully someone has done a test verifying a probes' accuracy with a CMM over time and can give some good input on this.

[Delw]

the haas is new, its a pretty tight machine, the 40-20 fadal is pretty tight also as its been rebuilt,
Circles on any of the 3 fadals I have owned and the many I have run, best I could ever get is .002 roundess on a 3.250" circle using a dial bore. the haas's that I had outside shops do the same part years back cause we were busy was within .001 or better. thats almost lathe quality I always been pretty impressed with the haas on circular interpulation. fadals have a tendacy to leave a small mark in the 4 corners when changing directions, the haas doesnt, the only small deviation is when I dont take into account the tool pressure on the start cut or not allowing the right amount of lead in and lead out. the circles are what I am trying to get at with the probe for inspection. back lash in a ball screw also needs to be considered.
when checking a length on a inspection plate with an indicator and joe blocks you dont have to worry about back lash, but on a machine with a probe and the ball screw nut moves both ways you have back lash

when I started out and still to this day with new customers we are requiered to have a seperate inspection room or we just dont get work. They actually come out for yearly surveys. I will have to ask one of my customers what they think about using the haas for a ccm and sample inspections. my cutomers do require me to get our inspection equipment recalibrated at least ones a year. Like our master blocks height guages inspection plates etc, we are allowed to do our own inhouse calibration of our inspection tools providing we use our masters.

the more I think about it and talk about it I might play around with the probe between christmas and new years to see how accurate it really is. I got a few dead nuts ring guages as well and its on a new machine that doesnt hawg steel or ss.

BTW I never said I wouldnt trust my mills with anything less than .005 -+.001 is pretty common of the parts we get and we have no problems holding size with the exception of a resolver going out., I said I wouldnt never rely on a probe on a machine to check anything less than .005 for inspection purposes. kinda like a mic to a pair of calipers.



Delw

[Matt@RFR]

Del, I'd be very interested in what your customers have to say about using the spindle probe for inspection. As long as the probe is calibrated to a calibrated ring, it seems that it would fall under current specs.

I completely understand your mic-to-caliper analogy, however in my case, if all you have is calipers, that's what you use. (I do have micrometers, just to be clear. )

[ParkerMillguy]

[QUOTE=Matt@RFR;705241]Del, no offense intended here, but how beat up are your machines that you won't trust them to anything less than +/-.005"??
QUOTE]

LOL...My thoughts exactly! However I would NEVER use my Renishaw system on our VF1 as a CMM. NOT bc of the accuracy of the machine (It's very accurate). I just don't trust the muscles of the operator clamping the work-holding scenario. You can have spring-back, warpage, etc. once the part is released from it's work-holding scenario. Don't you want to verify quality in it's free state?
I work for an aerospace facility and our probing system is very valuable as I can reset our position to a known starting point (Fixture location) and then probe a feature to adjust for a misassembled component. I will set a true position variable, and if it exceeds that limit, the machine will go into alarm so I don't waste time cutting something that is already junk.
It is also valuable to update offsets for cutter compensation as tooling wears (The machine never mis-calculates and adjust the wrong way!). I can't remember the last time I used an edge-finder. Probing makes set-up very quick. I now fixture parts for the best work-holding scenario (Not for Datum-seating) as I can probe every part for it's position...the operator just has to have it loaded within a 1/2" of it's position!
For value-added machining scenarios, you can't beat the probe. But it's not for post-process inspection IMHO.

[mls]

What makes the haas cut a round circle better than a fadal I have 2 haas mills and it surprises me how good they are and where I work part time they have a 1997 or 1998 vf3 which is beat up (they do no maintnace) but we had a job that required a .0005 tolorance on a 1 in (I think) bore the guy that progamed it used the g13 and repeated it once or twice and the bore checked good with gauges and the cmm.

What seems to be the difference

[Delw]

I talked to a few buddies and one inspector friend of mine about this.
the inspector and one lead machinest friend saw nothing wrong with using the probe for inspection purpose's, one of my buddies also runs a shop that I get work from says NO way, its not right as its not an inspection room its not a controlled enviroment and unless the machines been calibrated often( he was saying maybe every 3 months cause of the heavy cutting on it).

The inspector did say this,( I dont do work for them he's just an old friend) he said. sure you can use it to inspect parts but are you willing to risk your reputation on it? he was saying he knows guys that do there work don't due proper inspections on sample parts, they just write in the numbers on the inspection reports. he said that their parts are good but if they ever have a problem with bad parts they wont use them anymore. but he thought there was no reason why you couldnt use the probe to check your parts.

he also said that you know if your doing a profile and you 1st article one part complete and it checks out then you shouldnt have any problems because the profile will be all realitive to the dims. if one or 2 dims are in then all the rads and angles should be in as well cause its the same finish tool and same program.

Personally I didnt think the inspector guy would be all for it.

Anyone else have any thoughts on using a probe for inspection purposes or did anyone talk to anyone else about it?

Delw

[NWSERVICEGUY]

During set up of a machine with the probe system the probe is calibrated to a precision ring. If you take a different size ring and measure it with the "probe bore cycle" you can then look at the macro variable ( 188 ) and see what the machine actually measures it as. I have however found that precision rings are effected by temperature so make sure you account for that. You can also measure a known boss, or gage block and see how accurate the machine is measuring. For parts that have real tight tolerances <.0003 " you may be pushing it. However if you measure a known distance and are more than a couple tenths off you may need to have the machine actually checked out and or the probe recalibrated.

[Delw]

theres alot of varibles and this is what proper inspection takes out.
Heat of the part you just cut,
heat of the machine is also an issue, if your ball screws heat up that will change as well.
heat of the coolant also plays a huge part into the size of the part

thats why a Inspection room is always in a control enviroment.

When I worked in a shop years ago we used to run APU parts( auxilery power units) shrouds/seals were 36" dia titanium and .125-.062 wall thicken with some goofy ass profiles,even profiles we had to build/put together the tool while it was in the machine inside the part.

we always had to let the parts set in a free state over night before the inspectors would even check them. Stress in lots of materials combined with heat especially in tit will do some wild things. some of our first article pieces had to wait 12 hour before we check them. that means the machine stayed idle for 12 hours, unless we decided to rough them then do finish pass's after the 1st article was bought off.
Man I do miss running big thin stuff is was challenging

Delw

[christinandavid]

I agree with the general consensus, which is that a machining centre cannot compete with a dedicated CMM for accuracy.

Remember a classic example on a VF9 which was making a pattern of holes on a 700mm or so PCD. Hole pattern tied up to 0.04mm true position.

Checked on machine with no error messages. Put it in inspection and the CMM made the pattern out of round, which was not unexpected because the part itself had sprung slightly. The issue we found was a pitch error between the holes, which righted itself at the point the machine axes changed direction ie N, S E & W. Backlash was blamed, but I then pointed out that the job had been machined using the 'move onto positions in the same direction' function (I forget the proper term, sorry).

One rule of thumb I follow when working with CMMs is if they pass your job they speak the truth - if they make it wrong then they are not to be trusted!

DP

[Donkey Hotey]

It would help if everybody agreed what they mean by 'inspection.' I see you guys posting about tenths and I'm laughing. The only hope I have of accurately measuring a feature like that is something a conventional, 8" micrometer will fit around and maybe bores up to 4". Beyond that, I can't measure it.

If somebody has to inspect to tenths, then an optical comparator or climate controlled CMM is a requirement anyway.

It all depends on the application. If I'm trying to verify that two mounting faces, on opposite ends of an aircraft bulkhead are square to another datum and POM 0.005" to each other, the machine is perfectly adequate to check that (and the only way I could measure it anyway). In a large part like that, material shifting as it's machined and the flexible nature of the thin webs are what will blow the final dimensions more than the machine could shift anyway.

[Delw]

It would help if everybody agreed what they mean by 'inspection.'.

Donkey your correct, I myself got confused which is easy to do.
After reading some and talking to some, I think it would be safe to say that you can use a probe to inspect parts on the the shop floor with no problems and in some cases use to it to parts for final inspections ( like if you dont have a CMM or Comparator and the parts arent to tight.)
for any mill work I dont think there would be a problem, for lather work then checking it on the mill there probaly wouldnt be a problem in most cases.

Delw

[gpcoe]

Depending on your customer, i would imagine that you may need to have regular alignment and ball bar tests to prove accuracy and maintenance of your machining center to be able to qualify it as a calibrated measuring tool. Calibrating to a ring gauge only tells the machine the offset of the ball center to the spindle center and the ball radius of the ruby. The limited non-official measuring I've done with the probes I use tells me that bore diameter measurement is accurate to roughly +/-.0003 and length is about +/-.0005. The bore I measure daily with repeatable results. The bore has a .0002 tolerance from a previous lathe op that I have to center on for my milling op and I monitor the bore size for miss-hits from debris and what not. This machine probably hasn't been re-leveled, ball bar tested, or aligned since it was originally installed in '06. I say this as a token to the accuracy of the measurement abilities of the machine within the limited scope of the measurement software. We don't have strain gage probes so I can't say much about any kind of roundness measurement.

Point being, I'm sure you could fairly accurately measure parts on your machine. You certainly won't be able to make projection planes or anything like you would with a dedicated CMM but you can monitor feature sizing and relative positioning easily. I think the end reason it's not viable as a inspection device is that you have no proof that the machine can accurately make measurements. Most machine tools don't have linear scales to backup the ballscrew encoders anyway. Our customers require routine checkups (calibration) on all of our inspection tools including our CMM's and CNC CMM's and I doubt they'd ask for anything less if we were to try and use our manufacturing equipment for inspection purposes. That's why we don't do it (I've only experimented with it). It certainly can be done and (it is done). There was an article recently in Modern Machine Shop that detailed some of the issues with using Machining Centers for inspection. It's worth a read if you're serious about doing something like this.

Greg