[Georgie]

Thank you for all your kind remarks about my attempt at the indexer. I have a drawing of the adaptor I made which I can send in various formats and can also point anyone who is interested at the motor supplier and driver supplier. It would be appropriate to do that off forum so if you let me have your email address either via a private message or leave your email address here I will get back to you.

HI,
I'm interested to know if anyone has made any PCB's for the electronics and keypad for Steves project and can I purchase a set anywhere ?

I have made my 4th axis which I will drive using my CNC router but I also want to use it with a hand held controller on my bench mill.

It uses a 3" chuck driven by a Nema23 stepper using a ratio of 6:1 drive pulley.

Hopefuly someone can help.

Thanks

George

[Georgie]

I plan to build this indexer. It's just what I'm looking for.

I have already completed my 4th axis and have it running under mach 3 on my router machine.

I want to make it dual purpose and use it on my vertical mill. The gear ratio is 6:1. 10tooth on the stepper and 60 tooth on the chuck drive. Do I need extensive modifications to the code so I can run the indexer with my set up.

Thanks

George

[kwackers]

Hi George,

It should work fine. If you're running a half step motor then that's 6*400=2400 steps per rev.
So just enter 2400 in setup ("Set Steps per 360").

Cheers
Steve

[Georgie]

Hi George,

It should work fine. If you're running a half step motor then that's 6*400=2400 steps per rev.
So just enter 2400 in setup ("Set Steps per 360").

Cheers
Steve

Great, thanks Steve. Methinks it's a trip to Maplins this weekend, they have some pre-sensitised copper clad board so I might have a go at making the PCB.

George

[allaces]

Steve,

I'm looking for a stand alone handheld controller like this for indexers I sell. Would you consider making them for money?

Please respond to [email protected]

Thanks,
Ray

[allanwinks]

Hi all
I am completely new to CNC and have started to build a router when i came across Steves wonderful digitiser. I have read my way through all the posts and wondered if anyone has a comercial PCB for sale? I have made them before but my success rate isnt high.

[Georgie]

Does anyone have a spare PCB for the Rotary indexer, failing that did anyone use Mach3 to cut one and would they care to share the code?

Many thanks

George

[Me2]

G’Day All,
I have been running my rotary indexer on my Rotary Table (ratio 90-1) for the last 18 months with no problem. Needed some more power so upped the driver capacity and added on a larger power supply still running great.
I acquired a dividing head recently (ratio 40-1) and added on the stepper with a ratio of 1.75 – 1 through the timing pulleys. This brought the ratio up to 70-1. Setup the second profile with that number running in half stepper mode.
When setting up to divide say 24 divisions the screen read out goes – 0, 14.02, 29.02, 45.00, 59.02, 74.02, 90.00 etc. The same sort of thing happens in degree mode using 15 degree steps. The strange thing is if I change the worm ratio back to 90 it works perfectly. I am absolutely stumped about what to do. I’ve reset the factory defaults no change I don’t know what to do next. I’m using the Hex Code 2.11. Has anybody else experienced this problem? If you have, what did you do to overcome it? If all else fails I’ll try programing another chip and see if that makes any difference.

Cam

[gandalf1]

Change your belt drive to a different ratio, 90:40 and you should get back to the original setting.
John.

[arvidj]

G’Day All,
I have been running my rotary indexer on my Rotary Table (ratio 90-1) for the last 18 months with no problem. Needed some more power so upped the driver capacity and added on a larger power supply still running great.

Cam

Sorry, not an answer but a question.

What size rotary table do you have?

What size and torque motor did you have ... i.e. the one that appears to be too small?

What size and torque motor did you upgrade to ... i.e. the one that appears to be an adequate size?

And possibly an answer to you question. ... Look back thru this thread as I remember seeing a similar situation discussed and Steve explaining what is happening.

If I recall correctly the "display strange numbers" has to do with the internals software keeping track of "position" as one of 36000 discrete positions ... 360 degrees with .01 degree resolution ...

... the software doing integer math and not floating point math ... i.e. the concept of a value ... even intermediate values in any calculation ... having something other than zeros to the right decimal does not exist ...

... and some rations having "nice" numbers when integer math is taken into account ... i.e. 36000 divided by 90 is 400 ... a nice whole number that fits well with the integer math concept ... but 36000 divided 70 is 514.2857142857143 which is a number that does not fit well with the integer math concept ... and what you see is the byproduct of arrangement.

Arvid

[kwackers]

Sounds like you're using an older version of the firmware. If you trawl back through the thread you'll see someone else had a similar issue which I looked into.

Turned out there was a bug in the display routine - the position is correct, it's just not reported correctly.
I just checked on mine (to be sure) and with a ratio of 70:1 the position is correctly reported.

You could just ignore it, as I said the actual position is correct, failing that you need to reflash the chip with the latest firmware and it should all be good.

(I'm fairly sure that versions of the firmware that require you to enter the number of steps per 360 degrees work correctly, but as I said if you trawl back through the thread you should find the point the bug was fixed)

G’Day All,

When setting up to divide say 24 divisions the screen read out goes – 0, 14.02, 29.02, 45.00, 59.02, 74.02, 90.00 etc. The same sort of thing happens in degree mode using 15 degree steps. The strange thing is if I change the worm ratio back to 90 it works perfectly. I am absolutely stumped about what to do. I’ve reset the factory defaults no change I don’t
Cam

[kwackers]

... and some rations having "nice" numbers when integer math is taken into account ... i.e. 36000 divided by 90 is 400 ... a nice whole number that fits well with the integer math concept ... but 36000 divided 70 is 514.2857142857143 which is a number that does not fit well with the integer math concept ... and what you see is the byproduct of arrangement.

Arvid

This is correct, worth pointing out that using floating point wouldn't fix this since the motor works in integer steps.

So for example 70:1 with a 400 step motor is 28,000 steps per rev, which is 1166.66667 steps per 15 deg.

For the first 15 degrees the software will step the motor 1167 steps (+ 0.33 steps error or 99.97% accuracy). Then it will step 2333 times (-0.33 steps error) for 30 degrees and at 45 degrees it will step 3500 times (no error).
It'll repeat this pattern for every 45 degrees so in this case the error is never more than +/- 1/3 of a step. Chance are the mechanics of the table are worse!

In essence due to the mechanics of the stepper motor the software can be out by up to half a step at any position. It does however always move to the nearest half step and so that's the largest error you should have.

[PIETABC]

Hello

I am a new member of CNCZONE. from Belgium

I will start to make the steve ward rotary table.
But I d'ont have the compleet source program
I need it because I want to change some things for my aplication .
I use on my PC Hi-Tech compiler.
My member name is PIETABC

Thank You ALL

[Me2]

Thanks for the replies guy's. What you have described makes perfect sense. I agree with the display being incorrect but the actual angled movement is spot on. In the mean time I've ordered a 36 toothed pulley to go onto the input shaft of the dividing head to bring the ratio back to 90:1. To close the loop I'll go back over the posts to find the latest code and re-flash the chip.
Thanks again for all the help.

Cam

[kwackers]

Thanks for the replies guy's. What you have described makes perfect sense. I agree with the display being incorrect but the actual angled movement is spot on. In the mean time I've ordered a 36 toothed pulley to go onto the input shaft of the dividing head to bring the ratio back to 90:1. To close the loop I'll go back over the posts to find the latest code and re-flash the chip.
Thanks again for all the help.

Cam

I think you want to be looking at around page 20.

It's a constant source of annoyance that I can't go back and edit old posts otherwise I could simply keep updating the first post to put the latest code up there (and remove some of the intermediate versions that are hanging about).
When I get a chance I'll get my website back up and running and keep everything on there.

90:1 is a 'nice' ratio, mainly because each (half) step of the motor is 0.01 degrees, so there are never any 'funny' numbers.
However, it works well with any ratio since the error is always tiny and well under the sorts of errors you'd expect in a typical table.

[Me2]

Thanks Steve,
The latest Hex Code (2.12) is post 128 for anyone else looking for it.

Cam

[joegib]

The very latest firmware is, I believe 'Rotary2.12c.zip' linked at Post#419 (Page 35). This version is thought to work for both the 18F452 and the 18F4520. However, if you've already burned v.2.12 to an 18F452, this further version won't provide any more functionality so there's not much point in using it (see Steve's note).

Joe

[hhogendoorn]

Somebody (Georgie) was asking for a spare pcb.
I have a spare double sided pcb for only the controller part, 18F452 included (programmed with 2.11).

See included pictures.

Not included :
MULTICOMP - 4244A1 - KEYPAD, 4X4 ARRAY
Steppermotor electronics.

Hans

[hhogendoorn]

Forgot to post the top.
Hans

[chucketn2]

I origionaly posted as chucketn. Password got broke and I can't get a response to fix it. So, I re-registered with a different e-mail.
I had trouble making a pcb to Steve's origional art. I tried Locas' version but could not get it to print to correct scale. That may have been the fault of the printer I was using at the time. Fast forward 18 months or so, a computer change, printer change, and loss of the portable hard drive that had all my notes and files collected over the last 10 years...
I Tried again. The same results with origional artwork. I downloaded Lucas' improved version and woopee!, it prints correctly. Now retired and no longer having access to laser printers, I will try Office Depot to print artwork for another iron on transfer try.
Thanks to everyone for keeping this project alive.
Steve, I hope you resurect your website soon.

Chuck (formerly chucketn)