Creeping Z

[rhfurniture]

I have a weird problem involving the very slow rise of Z in the course of carving a name (12 letters 18mm high).
This is something I do very rarely, so I am happy to use Part Wizard to prepare the file, and as far as I can tell, there is no problem with the file itself, though it is far worse when the lettering is Times New Roman, where the Z ends up about 1.5mm higher than when it started, rather than Arial, where it ends up by 0.2mm. Shopbot has performed flawlessly in it's Z up to now.
I have checked, cleaned and re-set the Z axis, pinion, springs (tried with both a little too much up-pull and a little too little), I have tested it with 200 up-down movements in both Jog(30mm) and move(0.5mm) where it behaved perfectly. The cutter can go no higher in the collet (and there is no force on it anyway - it is a brand new Armana RT 90deg) and I have tried setting ramps to both default and something slightly gentler, and I have slowed movement down to 20mm/sec (10 in Z).
I have upgraded my computer's USB from 1.1 to 2.0 with a card and tried a hub as well. I have V3.4.30 software running in win2K with a 4G prt.
I think the USB2 helped to get the Arial OK, but the TimesNR still had this effect.
I have done the job now (in Arial) but would like to know what the trick is so I can use Times New Roman in future, and I also need to know that my Z is reliable (even though it passed my updown 200 times test with flying colours).
I would welcome any suggestions or experience of this issue.

TIA, Ralph.

[rhfurniture]

I have just tried the Times file as an aircut, and it ended up with its Z at 0.9mm higher than when it started.

[Brady Watson]

Ralph,
Check to make sure that your Unit Values, ans specifically your Multiplier values (under VR) are correct.

-B

[rhfurniture]

Brady,
Unit Values are all the same (X,Y and Z)at 90.2296 (I am metric). Given that all else seems fine, and I have the same prt motor/gearboxes all round, that should be OK. The multipliers are all set to 1. I do not understand what they do, and as I have a 4G prt, not sure if they do anything.

R.

[bill.young]

Hey Ralph,

When you say you slowed the speeds down, did you do that in the ShopBot software using the MS command, or did you change it in the tool database for that bit in PartWizard? You might look in the part file and see what PartWizard is setting the speeds to...those are the speeds that the file will run at no matter what the Control Software is set for.

Just a thought,
Bill

[rhfurniture]

Bill,
No I checked it in the file, as well as on the screen while cutting the file. I have just re-run my up/down move 0.5mm (at 20mm/sec) but this time for 500 lines (ie 250 cycles) test. Ended exactly at the same Z. Part-wiz default speeds for this bit were 20(x/y) 10(z). Is this about right for lettering work? (I do it so very rarely)

Best wishes, R.

[rhfurniture]

If someone could email me the speedtest.exe utility I would like to check the USB (it did not come with the software version I have, and I cannot find it on the website).
ralph@rhfurniture.com (mailto:ralph@rhfurniture.com)
Thanks again.

R.

[benchmark]

Hi Ralph,

I tried sending the speed test from 2 email addresses.... they were both bounced ?

Paul

[rhfurniture]

Hi Paul,
Works for me. Maybe uk2 or tiscali (it goes through both) bounces exe files for virus - zipping may work. Alt send to:
rhfurniture@blueyonder.co.uk (mailto:rhfurniture@blueyonder.co.uk) (I'm there now)

Thanks,
R.

[Brady Watson]

Ralph,
Your Unit Values (not VR, but VU - made a mistake above) - should not all be the same UNLESS you are running the same exact size pinion on your X,Y & Z axes. If running a PRT (non-benchtop) then you should have a 20T pinion on the Z and 25T pinions on X & Y. Are you sure they are all the same?

-B

[rhfurniture]

Brady,
Yup, all 20T - I have a benchtop.
R.

[rhfurniture]

Now FIXED. - Thanks Scott (shopbot).
Upgraded to software 3.5.3 and also did the firmware and it now works fine and smoothly.
Apparently the version I was using (3.4.30) had step loss issues (though I never realised it till now...)

R.