Denford Novamill troubleshooting help

Hey guys I’m looking for some suggestions on troubleshooting my issue further.

I purchased this cnc with unknown issues. The x axis stepper was jittering so I swapped the output from the x and y. The same stepper works on the y axis so it doesn’t appear to be an issue with the motor. I don’t see any visible damage to the control board - any suggestions on what I could test with my fluke or something to narrow down the issue? Thanks!

Hi Drew,

Have you searched on the Denford forum, they have a novamill section **LINK**

Pete.

The Novamill is an excellent machine mechanically but the electronics is long in the tooth now. I have I think a full set of the electrical drawings from the Denfordata site but they don't include the actual schematic of the control PCB. It sounds like there is a problem with the stepper driver, and if it's possible to get access to the step and direction signals you could easily replace that with a new unit. Hopefully someone on the Forum could advise on if/how this can be done. I doubt there is any useful troubleshooting that could be done with a Fluke DMM.

I acquired my Novamill without the electronics and built my own 4-axis control box using more modern drivers (this was 10+ years ago - better ones available now), so I don't have any knowledge of the original system. A good UK source for CNC bits is

**LINK**; they could supply a complete controller box but it's about £800. I built my own 4-axis one for ~£200.

there's also these people:

**LINK**

for steppers and drives.

Assuming you do have the Denford electronics that could provide the PSU and spindle drive and an enclosure. It's worth upgrading to a modern controller too such as Mach4 or UCCNC. You could choose Mach3 - I still use that but it hasn't been maintained for 7 years or more and given they still charge for it it's daylight robbery.

Or if you fancy an adventure there's always LinuxCNC...

I've built CNC controls for the Nova and my lathe and a 2-axis one for big mill + dividing head, happy to share info and help if I can. Where are you located?

Hi, I maybe able to help with a driver if I recall correctly they were digiplan?

If you can send a photo that would help.

Barry

The motor may be being driven too fast and is skipping steps.

I attached a picture of the board and also made a post on Denford’s site. You should be able to zoom in to view the chip info. I am located in St. Louis, Missouri, USA. If I go any route from here I guess it would be the LinuxCNC route. I guess that means the driver board is proprietary hardware?

https://imgur.com/a/kaMY0wJ

https://imgur.com/a/Ek7t0tf

thank you for your replies

Edited By Drew Montgomery on 21/09/2022 17:36:56

Edited By Drew Montgomery on 21/09/2022 17:42:44

Thanks for that Drew. Looks like the board carries all 3 drivers and is based on the very old ST LM297 controller and LM283 bridge output. Vary hard to replace just one channel I think. Frankly if you are thinking of the Linux route I suggest you regard the existing electronics as a source of bits such as the PSU and spindle drive, and fit either a new breakout board and stepper drives if you have or can get a pc with parallel port, or better one of the few ethernet motion controllers that are compatible with Linuxcnc. The Linuxcnc forums should help.

If you stick with windows you have a much wider choice of either ethernet or usb motion controllers and can use Mach3/4 or UCCNC or other packages under Windows 10/11. On my nvm I use the uc100 usb controller with Mach3 on a Dell Win10 mini pc.

By the way those old drives can only do full or half steps while modern drives can do usually up to at least 1/32 microsteps which makesnfor smoother operation and higher speed.

Hi, I am sorry I can’t help you with the boards I have as they are single channel boards and probably even older than yours.

That said I agree with John H’s suggestion that in the long run you might be worth while converting over to Mach or Linux and getting new steppers, drives and controller. I personally would go with Gecko drives but there are plenty alternatives out there. The steppers used by Denford were good at the time but the newer motors tend to have more torque.

If you want to try and keep the original board I would check the diodes, L297 and L298N on the axis that’s causing problems. The easiest to check with a standard DMM would be pins 2 and 12 on the L297 (GND and +5V) and 9, 4 and 8 on the L298N (5V, 36V and GND) with the power on. Other things would need an oscilloscope. Other things you have checked by switching the motors over. To me it sounds like one of the sides on the L298N is not working right. Try putting the meter from each motor lead in turn to earth when the drive is energised and measure the voltages?

I hope this is of some help

Barry