3.7 mr3 true position problem

[rebeldude]

I am trying to get the true position for a hole that is on a different depth of the sheet metal part. Datum A actually is set from 2 points on one depth and 2 points on a different depth. The hole is on a different depth completely other than the datum A's. When I try to tye the true postion in for this hole, it actually comes up with different numbers than the "x" and "y" and "z" location actually is. Has anyone else had this proiblem and if so, how did you remedy it?

[Wes Cisco]

I am trying to get the true position for a hole that is on a different depth of the sheet metal part. Datum A actually is set from 2 points on one depth and 2 points on a different depth. The hole is on a different depth completely other than the datum A's. When I try to tye the true postion in for this hole, it actually comes up with different numbers than the "x" and "y" and "z" location actually is. Has anyone else had this proiblem and if so, how did you remedy it?

Sheet metal you say, so circle, not cylinder? Do you have the "use datums" box checked in the true position window? If you are looking for true position of x & y axis, than the z's just plain do not matter. Establish Datum A in whatever way, then leve to Datum A.
One thing you might check is the perpendicularitly/paralellism of your cylinder/circle to Datum A. Are you using surface hits? Just some thoughts. HTH

[dccfreak]

Rebeldude,

Is it possible for you to use a CAD model? This is what I usually request when those sorts of thing come up. Otherwise, I will use jackstands, fixture plate standoffs, or whatever I can find and set them according to the various depth differences and use them as datum targets with the part resting on them at their respective locations. I usually use the fixturing plate to establish my alignment level ultimately, because this most closely represents an assembly environment IMO for a stamping. This way, the actual high points on the surface of the part will contact the standoffs in the same fashion as in the assembled state.

Then I will just set a point to correspond with the Z origin point depth on the print from one of my standoffs. You should be able to then measure your hole and get a pretty accurate result of where your hole is. Sometimes I will level on the surface around the measured hole also if it is on a plane that is somewhat irregular to the datums and then just recall my alignment before measuring.

I have used this many times in the past for automotive parts and it seems to work pretty well for me and I am also able to correlate with other gages. Ultimate ease would be to have a CAD and use an iterative alignment though.

Hope that helps a little.

Brad

[tking]

Reb,

Did you create an offset plane to estalish the -A- plane?

TK

[rebeldude]

Hi everyone,
yes I am using the check box "use datums"

I am using a cad model and an iterative alignment

No I didn't create an offset plane to establish the datum a . It is established from the cad and actual points of course

But you got me thinking!!! I do that sometimes. The datum A points are not flat to the cad. There is a variance of up to about .30 mm around the 4 datum A points from the cad,because the part is not parallel like it should be. Do you think this is causing the problem of a difference between the location numbers and the true position numbers?

Again, thank you all for your help and suggestions!!

[Matthew D. Hoedeman]

I will not use datums when doing a TP. If I need some other alignment, other than my CURRENT alignment, I will make one that uses the datums called out, then just do TP, no datums. The way Pcdmis does TP to datums will screw you up every time. Instead of adding the datum MMC to the feature you are dimensioning, it will MOVE the datums around by the datum MMC bonus to make the hole you are trying to dimension as close to zero as it can get. I think it should show you HOW MUCH it is off to the datums and add the MMC bonus to the TP tolerance. HOWINTHEHELL are you going to figure out how much and which direction the hole would actually have to move if Pcdmis is moving the datums around?

[Wes Cisco]

WMS What Matt Said, as usual.

[rebeldude]

Matt,
I want to make sure I understand. I should just leave the "use datums" check box unchecked?

[#2#]

Matt,
I want to make sure I understand. I should just leave the "use datums" check box unchecked?

Do an alignment to the part using the datums necessary for the callout. Then dimension leaving the "use datums" box unchecked. Select only your feature you want the TP of. Click OK. After the TP dim recall a different alignment if necessary.

[Matthew D. Hoedeman]

YEP, and ONLY select ONE feature and dimension it, as soon as you click on a second one, it ASSUMES you are using datums. It is not like reportin DIM-LOCATION where you can pick all the features you want to dimension at one time. You have to pick a feature, dimension it, pick another feature, dimension it, one at a time.

[rebeldude]

Great!
Thank you all very much for your help!

[craiger_ny]

These guys are all quite right. If you need datum bonus or just plain need to understand what the geniuses at Wilcox were thinking when they made PCDMIS act like a hard gage read this whole thread (if you have time of course).

That was a good thread that addresses what you have going on.

Craig

[tking]

But you got me thinking!!! I do that sometimes. The datum A points are not flat to the cad. There is a variance of up to about .30 mm around the 4 datum A points from the cad,because the part is not parallel like it should be. Do you think this is causing the problem of a difference between the location numbers and the true position numbers?

Reb,

I think as Matt explained that the Tru Pos monkey in the demon is your problem here. Were it mine to do I would create the -A- using the offset function, after that WMS.

Are we having fun today?!

TK

[rebeldude]

Thank you Craiger and TKing!. I'm proud to be part of this Forum with all of you!! Many of times you have all kept me from pulling my hair out!!

[Shelley Liebman]

I have had so many people ask me why I don't use the "use datums" option--this is a great example of why.
I always setup my alignment using the datums called out, then just position the hole or whatever......
I figure, once I have done my measurements, I can use as many different alignments as I need, and I don't have to worry about trying to figure out what pcdemon is doing with the "use datums" thingy