Hello Guest it is August 19, 2019, 02:32:45 AM

Author Topic: Spindle DRO not working in Mach3 Turn  (Read 8803 times)

0 Members and 1 Guest are viewing this topic.

Offline MHJ

*
  •  18 18
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #10 on: August 24, 2010, 11:17:02 AM »
Yes I do, erratically at 4 - 7 sec intervals at 100rpm, whereas the led on the input pin of the bob is flashing in sync.

Offline Hood

*
  •  25,849 25,849
  • Carnoustie, Scotland
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #11 on: August 24, 2010, 11:31:33 AM »
Just looked again and see you have an index debounce of 100, set that to zero and it may help.

Hood

Offline MHJ

*
  •  18 18
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #12 on: August 24, 2010, 11:44:18 AM »
 index deb0unce now at 0 and the dro is now constantly changing from about 35000 to 1100

Offline Hood

*
  •  25,849 25,849
  • Carnoustie, Scotland
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #13 on: August 24, 2010, 11:53:29 AM »
try increasing a bit then and see if it helps, all my machines were zero on index debounce when using the PP but maybe you have a bit more noise.
Hood

Offline MHJ

*
  •  18 18
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #14 on: September 01, 2010, 08:38:42 AM »
UPDATE:

I have discovered what appears to be the source of the problem.
The second parallel port (Moschip 9865) has around 3.5v on several of the pins, without any program accessing that port. This only came to light (huge pun intended) when yesterday evening the tube above the lathe failed and I could see that the LEDs on the Homann BOBs were glowing very faintly...yet the power supply to them was not energised. I disconnected the LPT2 plug and the leds all went out.
These parallel ports are, as far as I know, the same as sold by CNC4PC, (I sourced mine here in the UK) and I have contacted Moschip to see if there is a way to configure this port so that there is no voltage on it when no data flow is present.
I moved the index pulse over to LPT1 pin 15 and hey presto! spindle true dro is working. I'll report the Moschip answer, as and when I have it, unless any of you Gurus already know what can be done !!!!
Thanks for all your input so far,
John.

Offline Hood

*
  •  25,849 25,849
  • Carnoustie, Scotland
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #15 on: September 01, 2010, 09:05:53 AM »
Is there any way you can connect the index direct to the port rather than going through a bob?
Hood

Offline MHJ

*
  •  18 18
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #16 on: September 02, 2010, 10:18:00 AM »
Hello Hood,

I don't think so.......

But I have heard back from the Moschip people:

"To show 0V on parallel port pins, parallel port transceiver need to be used which is added in most of desktop Machines.
Use of Parallel port transceiver is optional, in case of MCS9865 – parallel port, board manufacturer do not prefer to use parallel port transceiver (Ex: SN74LVCE161284)  to show 0V when parallel port is not in use."

So what I need now is a parallel port card which has this chip (or an equivalent) on it.

Ho Hum eh? Regards, John.


Offline simpson36

*
  •  1,374 1,374
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #17 on: September 03, 2010, 12:29:19 PM »
From the FWIW department, I have some similar hardware and had similar problems.

I am using a Roswell dual port PCI add in card. Not sure which one at the moment , but it uses a MOS chip. I posted the exact card model here a while back. Note that the drivers that came with the card would NOT work with Mach, but the drivers from the MOS site (for the specific chip on the card) work perfectly.

I have a CNC4PC BOB (G1) and also the Homann BOB that you are using. I was never able to get the Speed DRO to work with the Homann board. I reported this to Peter early on, but never got any response. The DRO works fine with no changes other than move the wire to the CNC4PC BOB . . which has a special 'high speed' input for that purpose.

Hope this helps you. 

Offline RICH

*
  • *
  •  7,351 7,351
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #18 on: September 04, 2010, 08:13:56 AM »
Quote
drivers that came with the card would NOT work with Mach, but the drivers from the MOS site (for the specific chip on the card) work perfectly.


Thanks Simpson, as i had the same problems with a few cards a long time ago. The drivers first tried were on the installation disc and followed the instructions. Actualy someone  sent me a driver that worked. Just should not need to do this kind of stuff.
My answer to this kind of stuff goes like this:
Buy the card but make an agreement ( before purchasing it "localy ") that full refund will be given or the seller will make it work,
If they won't do that ...........walk away and go somewhere else.

RICH

Offline MHJ

*
  •  18 18
    • View Profile
Re: Spindle DRO not working in Mach3 Turn
« Reply #19 on: September 05, 2010, 08:56:12 AM »
Thank you all for taking the time to add to this enquiry.

I feel that soon I will be in a Home For Distressed Machinery Retro-fitters. I have considerably less hair than before and I've smoked more fags in the last fortnight than in the last six months.
(just to clarify that last statement; "smoking fags" if you are East of the 45th parallel refers to the consumption of cigarettes and not some sort of mass-murder, or worse.)
I've reinstalled LPT2, with the correct drivers, port address is still 1040, and pins 2-9 as inputs box is checked. I've checked in BIOS and in hardware devices, all appears OK. On the LPT2 Homann MB02-V6 board, the jumper is in the correct position for pins 2-9 to be inputs, but these are not seen by Mach3. I changed over the MB-02 LPT1 to LPT2 and vice-versa, still no inputs seen on pins 2-9. The normal inputs/outputs work on both BOBs and are seen by Mach3, so I see no issues there.
So is the lack of inputs 2-9 due to the LPT2 card or the Homann BOB? Or a combination of the two?
Well, signals on pins 2-9 are going down the LPT2 cable and reaching the computer, I can see these with my multimeter, so the BOB can be ruled out.

I contacted these people here in the UK: http://www.brainboxesstore.com/product_details_25.htm  
One of their hardware Engineers got interested in my enquiry and very kindly put one of their cards on an oscilloscope, checked all the I/O's and here's what he said:
"I think it is very likely that the UC-146 will meet his requirements. The chip we use for the UC cards drives pins 2-9 (the data pins) of the LPT port low at reset, and all PCI-compliant motherboards hold the PCI slots in a reset state during power-up.  So by design, the UC-card LPT ports operate in the way he requires. I did some oscilloscope measurements with a UC-146 card, and I did find that there was a small glitch on these pins at startup: there was a pulse of up to 0.88V lasting for 360us.  This is probably occurring as the power to the card ramps up, and has to reach a certain point before the logic in the chip starts working and drives the signals low.  The pulse length may vary depending on the host PC, but I guess it would not get much higher in voltage. When I set the 'scope to trigger at 1.0V, it did not trigger on any of the pins 2-9 at power-up. That covers the behaviour of the hardware, but might the Windows device driver for the LPT port cause the pins to change?  I haven't tested this, but I doubt it because we use the standard Microsoft device driver for our LPT ports, and I get the impression that the customer is already using a standard LPT port (which will use the same driver) without problems."

These cards are not cheap at 57GBP + tax+ shipping, but if it preserves what is left of my sanity I'll be more than happy.........watch this space.