Rotary table indexer

Quote:

---

Originally Posted by Ben Olson

it looks like I have some checking to do , I did verify the chip ( using IC Prog ) but I was unsure about the 10k variable resistor ( I am an electronics newbie )

I have to turn the pot all the way one direction to see the squares

the power supply to all the pins on the IC seem to be OK

I am a little unclear on checking the config bits is this the same as verifying?

thanks , Ben

---

Yes, that variable resistor is controlling the contrast (I beleive) on the display, you should turn it down till you can not see the black squares. (I think. LOL! really helpful huh? :)

RE the config bits, there will be a way in your programmer program, the software that programs the pic chip, to configure the configuration bits.. there will be things like ext osc, int osc, memory protect, etc.. you should be able to read the pic processor's memory space out and those config bits will reflect how the device is programmed. You need to make sure that the configuration bits are set correctly according to the details of the project plan. I'd have to dig to see it again, but there will be , SOMEWHERE , in the design, an indication as to which configuration bits need to be set and which ones need to be cleared. Most often, the project, when compiled or assembled will have a configuration word in there thta will have all those details specified, and the programming software will pick it up from there. but if not, you have to set them manually in some kind of menu before you program the device. Also, if they are in the project itself, you will want to make sure tht your programming software knows to use the ones in the project files and not the ones that you select manual, or vice versa if they are not set within the source code file.

I hope this helps.

Rick

everything is helpful as you fellows know a LOT more about electronics than I do

I'll keep plugging away ( seriously considering re doing the board at this point ) and will report back with my findings....

thanks

Quote:

---

Originally Posted by Ben Olson

everything is helpful as you fellows know a LOT more about electronics than I do

I'll keep plugging away ( seriously considering re doing the board at this point ) and will report back with my findings....

thanks

---

hey there ben..

I checked again, looks like you will need the HS oscillator configuration bit set to make it run, go back a few posts and check what kwackers wrote.. As I do not see any configuration data in the hex file. So you have to set the config manually, which means, at this time, your PIC is most likely just set up wrong..

Keep pluckin', ducky!

Horsedorf

Hi Ben,

I wouldn't bother redoing the board - unless it's iffy and doesn't look like it would work :)

As Horsedorf says the configuration bits are the key. In your software there'll be somewhere to set them up.

There are 3 main config bits that require setting, if any of them are wrong your device won't run and the behaviour will be as you describe.

These are.

Oscillator - set to HS-PLL.
Low voltage program - disabled.
Watchdog timer - disabled (may say - enabled by SWDTEN bit).

Most other settings are unimportant - you could just set them to disabled were possible.


Since this should be the easiest to try - I'd make sure the configuration is correct before doing anything else. Shame to break something that is working.
(Check pin 1 of the PIC is at 5v too - if you read back earlier up the thread someone missed the connection for that on the circuit diagram....)

For anyone coming late to this thread - all the latest stuff is available on this post:-

http://www.cnczone.com/forums/showpo...5&postcount=45

What program do you use to program the chip? ICProg has the configuration all FF out when you select the 18f452 chip. ICProg has a poor verify when used. WinPic is fast, verifys okay, but when verified using ICProg is poor. I used PicPgm which is slower, but it verifys good and it also verify good in ICProg. I am using the JDM programmer as well. I also only get 2 rows of squares. I even started another board with only the power supply, oscilator, IC and the display installed as to eliminate any problems and still only get the 2 rows of squares. Any suggestions to try?

Hi, not sure why the config bits aren't set in the original file. You can of course just set them as described earlier.

However attached to this is a .hex file with config bits set for V2.0 of the software.

If it you still have problems please check your setup as described earlier in this thread.

There, that fixed it. Thankyou.

David Paulson

A couple of bugs I found.

1. The ratio. Setting the ratio to 60 and then jogging +0.01 the display says 0.015. It seems that it is using the 90 that is stored in the 'system default'. for the pulses to move while the display is using the user selected 60 to calculate the amount moved. This happens with several different values, but if it is a multiple of 90, it seems to work right.

2. If you ground the sense pin twice, it will store the second 'click' and execute the wait for sense again. I think any 'click' should be disguarded and only accept one 'click' when on waiting for sense.

How does the acknowledge work in the program mode?

Overall, this is a very cool device.

I felt a need to post just to express my appreciation for all the obvious effort kwackers put into this and his generosity at posting it and giving assistance.

This is the first time I have joined a discussion board but I was compelled to by my excitement at finding this thread. I am a newbie at home machining and only "book-learned" about the CNC aspects of the craft. I have been collecting equipment and tooling for my small mill and lathe and recently reached the point that I was considering some type of rotary table/dividing head. I was evaluating the manual option while at the same time thinking they all looked like a lot of work to use, feeling that "there must be an easier way".

Not long ago I stumbled upon a simple looking electronic dividing head on youtube (http://www.youtube.com/watch?v=V2NOHWKqqi8#) but I had, until now, been unsuccessful at figuring out how to make one. I bought a couple books on electronics and using PIC but I have been frankly slogging thru trying to get enough of a working technical understanding to make a stab at making my own controller.

Then (drum roll...) I found this thread. You can perhaps imagine my excitement.

Anyway, I am going to dig into the books and start gathering components to take a swing at this. If anyone has any advise they think would shorten the learning curve I am all ears.

Looking forward to participating here in the future.

Thanks.

I sent the parts list to Mouser, Alllied, Newark, Digikey and MCM.
ALL of them told me I needed to provide Catalog Part Numbers in order for them to quote and/or fill an order. Is it unreasonable to think these inside sales people at these companies should be able to look up their own part numbers in their own companies data base and provide ME with their catalog numbers? Or is this another case of lazy people that don't really want to work for their pay? I thought their -job- is to quote price and delivery on component lists and cross component part numbers to catalog numbers! My job is to give them my money so they can get a paycheck at the end of the week.....not spend an hour with my nose in their catalogs.....

rant off

IF someone has or develops a list of part numbers from one of the USA electronics supply houses, would you consider sharing it with me or perhaps in this thread for the benefit of all....??

Thank you
Dan K

I had some circuit boards made up. They are about 4 x 3 in size. The only problem is that the display pinout is flipped, so you need to use wires to connect the lcd instead of the nice though pins that kwackers shows in his assembly. I have about 8 left. My cost was 30.00 CAD ea. Anyone who wants one, I can send cost + postage.


David

Just send me a PM or email to [email protected] as not to contaminate this thread.

Thanks

I'd be interested........too bad the pinout is wrong.

Hi, apologies for coming late to this - for some reason I no longer get alerts when someone posts on this thread.


With re bug 1.
I'm not completely sure what you mean.
However, I'll explain how it works, if this doesn't agree with what you've found let me know and I'll have a look.

With a 90:1 ratio and a half stepping motor, the minimum the motor can step is 0.01 deg. Therefore pressing the 0.01 jog button will always move 0.01 deg.

With a 60:1 ratio, each step is 0.015 degrees. The device can no longer step in 0.01 deg steps, so it defaults to stepping one step - hence the 0.015deg steps.
Obviously there is no fix for this - however if you have a microstepping driver you may want to use that and change the ratio to compensate.

Bug 2. Agreed, it shouldn't store the sense. I'll look into that.


In program mode there are two commands to do with sense and acknowledge.
First is 'wait for sense' which obviously just sits and waits for the sense line to change.
Second is 'acknowledge' which sets the acknowlege line. These two commands would normally be used in pairs to provide the correct handshaking.

Their main purpose is to link two units together (a friend of mine built a automatic clock gear/pinion cutting machine using a pair).
However they can be used with a PLC for production automation or any system where some other control system can manipulate the control lines.

Hope this helps.
Steve.

Quote:

---

Originally Posted by dpaulson

A couple of bugs I found.

1. The ratio. Setting the ratio to 60 and then jogging +0.01 the display says 0.015. It seems that it is using the 90 that is stored in the 'system default'. for the pulses to move while the display is using the user selected 60 to calculate the amount moved. This happens with several different values, but if it is a multiple of 90, it seems to work right.

2. If you ground the sense pin twice, it will store the second 'click' and execute the wait for sense again. I think any 'click' should be disguarded and only accept one 'click' when on waiting for sense.

How does the acknowledge work in the program mode?

Overall, this is a very cool device.

---

Can you not just mount the LCD on the other (i.e. component) side of the board?

Quote:

---

Originally Posted by dpaulson

I had some circuit boards made up. They are about 4 x 3 in size. The only problem is that the display pinout is flipped, so you need to use wires to connect the lcd instead of the nice though pins that kwackers shows in his assembly. I have about 8 left. My cost was 30.00 CAD ea. Anyone who wants one, I can send cost + postage.


David

Just send me a PM or email to [email protected] as not to contaminate this thread.

Thanks

---

Just to re-iterate, none of the components are critical and all should be easily obtainable.

When selecting components the only thing to concern yourself with is the physical size - if you can squeeze it onto the board it'll be fine.

The only component to be careful of is the 18F452, these are available in a number of different packages. Make sure you buy the 40 pin DIP (look at the internal photo if you're unsure what it should look like).

Quote:

---

Originally Posted by kmcwhq

I sent the parts list to Mouser, Alllied, Newark, Digikey and MCM.
ALL of them told me I needed to provide Catalog Part Numbers in order for them to quote and/or fill an order. Is it unreasonable to think these inside sales people at these companies should be able to look up their own part numbers in their own companies data base and provide ME with their catalog numbers? Or is this another case of lazy people that don't really want to work for their pay? I thought their -job- is to quote price and delivery on component lists and cross component part numbers to catalog numbers! My job is to give them my money so they can get a paycheck at the end of the week.....not spend an hour with my nose in their catalogs.....

rant off

IF someone has or develops a list of part numbers from one of the USA electronics supply houses, would you consider sharing it with me or perhaps in this thread for the benefit of all....??

Thank you
Dan K

---

Quote:

---

Originally Posted by kwackers

With bug 1.
I'm not completely sure what you mean.
However, I'll explain how it works, if this doesn't agree with what you've found let me know and I'll have a look.

With a 90:1 ratio and a half stepping motor, the minimum the motor can step is 0.01 deg. Therefore pressing the 0.01 jog button will always move 0.01 deg.

With a 60:1 ratio, each step is 0.015 degrees. The device can no longer step in 0.01 deg steps, so it defaults to stepping one step - hence the 0.015deg steps.
Obviously there is no fix for this - however if you have a microstepping driver you may want to use that and change the ratio to compensate.


Hope this helps.
Steve.

---

Yes of course, you are absolutely correct. I was just playing with the controller on the desk, and getting familiar with the programing when I came across this and never gave it much thought, at least not in the right direction, as to why this was happening.

Thanks again.

Thanks Steve (kwackers),

I supplied the 5V DC required for the “Rotary Table Divider” circuit from the step motor driver card. And I did not use limit/control and sensor parts. Besides, in order to be able to make the circuit board smaller, I pulled the unused legs of the microchip out of the socket of the microchip. (See: Pht2). I also did not use 10K VR1. I connected the contrast leg of the LCD directly to the ground. The system the photos of which are attached works very well. Once again, thanks a lot for your efforts and contributions, Steve.

Best regards,

Erginer, from Turkey

I did

Jog CCW mode not working correctly. Press any keys for the CCW and goes 0.1 CW

software http://www.cnczone.com/forums/showpo...6&postcount=67

Fix :)

short-circuit limit2 with GND

Tanks