PDA

View Full Version : SSO/FRO (Spindle Speed/Feed Rate Override)



m_c
16-10-2016, 09:23 PM
I'm thinking how best to implement these.
I could use either encoders, or potentiometers.
.
My current thought is encoders, as overrides could then be reset in software, but when would you want to reset them?
I'd have the ability to program things to reset whenever I wanted, but when would that be?
When you disable the override?
When a new feedrate is called?
.
Then do you reset them when re-enabling the overrides?
.
At the moment, I rely on onscreen overrides which you have to manually reset to 100%, however I can think of situations where all the above could be handy, so I'm struggling to make decision.
.
Regarding the encoder v. pot, I'm thinking encoder, as there could potentially be several meters of wiring needed between the control panel and cabinet, so a high value pot would be susceptible to noise, which could cause unwanted fluctuations (off course, a small capacitor at the controller analogue input would help smooth out any noise).
.
So I'm interested to hear other peoples thoughts.

magicniner
17-10-2016, 12:46 AM
but when would you want to reset them?

What's your rationale for implementing this? I use overrides to optimise jobs, for anything other than a one-off I make notes and go back and correct the speeds & feeds in CAM, then re-post the code with the correct feeds & speeds.

- Nick

m_c
17-10-2016, 01:08 AM
Probably me overthinking the possibilities!
.
Ultimately the g-code file gets updated (whether that be direct edit or via CAM), but if I'm optimising to find the sweet spot, it may involve adjusting the override at various points during the job. So do I reset after each commanded speed change, or keep the override value and simply re-enable at that value?
.
With a Pot, enable the override, and depending on where the pot has been left, the override will jump to that setting. I could program things so it doesn't instantly move to that setting, but at some point the dial has to sync back with the setting.
Having an encoder allows me to do pretty much whatever I want with the last setting.

Neale
17-10-2016, 10:07 AM
Overthinking - now there's a man after my own heart :smile:

I would see two reasons for encoder over pot. I think that you've mentioned both. One is that at switch-on, if no other time, you can reset to 100% in software as the encoder is a relative rather than absolute device. No starting up at 200% because you didn't notice that the pot wasn't reset. Second point is about noise. Yes, you can filter low levels of induced noise, but my own experience is that it can be more problematic if the pot track itself is noisy. I built an Arduino-based temperature control for a soldering iron recently and suffer badly from this; I'll be changing to an encoder when I get time to do it. The software to support one or the other is pretty trivial. I suppose that another answer would be to buy a decent quality pot and not use the 30-year-old one that was already there, but I think that the first reason is enough in itself.

I can't make a call on other times that you might want to reset in software but switching on in a known, safe, state would swing it for me anyway.

Boyan Silyavski
17-10-2016, 05:16 PM
I don't understand exactly what you are talking about, but if you are talking and thinking about the chinese controllers, i have to clarify that-if the spindle is not a servo motor but HF spindle-then its best to make default frequency 10 000 and use SRO on panel. 2 benefits of that- its easier to adjust voltages on the VFD and thus the spindle speed at all speeds is more correct. Plus the override on the board i am testing is 10% 20%, etc, so 100 is easy to divide by 10 and usually 16000 rpm are programmed not 1575 rpm.
Hope that makes sense.

At the other hand its better to have an encoder. Especially if you don't like to change tools and do 2 or more different HSM operations with the same tool. Which will require to stop and change the tool and speed change. But if an encoder is fitted, then you could cheat by stop. change nothing but speed automatically from program and then continue with another/ the same actually/ tool. Recently i am doing quite a lot of that. Still the stupid way though.

m_c
17-10-2016, 10:33 PM
The controller would be a Dynomotion KFlop, so I have lots of flexibility in programming.
.
I still need to run through a list of inputs/outputs for this machine, as if I allow for how I ultimately want to upgrade this machine to, then I may only have one encoder input spare, unless I move some of the differential inputs to other inputs (I've got a Kanalog, so the differential encoder inputs can be used as standard differential inputs).
.
I've had a reply from the man behind Dynomotion, and he said either way would work fine, and if noise was an issue with a pot, just fit a bigger capacitor over the input wires, as you don't need that fast a response.
.
If I was to use pots, then I could have an option for it to either go straight to the pot setting, or not to move until the pot is returned to near zero.
.
Sometimes lots of options isn't good!
I've got a few other more pressing issues to deal with, so I've got plenty time to think about this.

JAZZCNC
17-10-2016, 11:30 PM
KISS every time. Why use expensive encoders when pots work.

Both SSO/FRO are invaluable to me for getting the best out of machine and G-code. Equally they have saved my bacon many times when I cock up in code and when working with materials like Ali they are priceless when hit hard soft spots which happens lots on cheaper softer grades.

Certainly worth taking the time to get right.

Boyan Silyavski
17-10-2016, 11:51 PM
KISS every time. Why use expensive encoders when pots work.

Both SSO/FRO are invaluable to me for getting the best out of machine and G-code. Equally they have saved my bacon many times when I cock up in code and when working with materials like Ali they are priceless when hit hard soft spots which happens lots on cheaper softer grades.

Certainly worth taking the time to get right.


I couldnt help but imagine this:


https://www.youtube.com/watch?v=6UfCv1XCMHg

JAZZCNC
18-10-2016, 12:14 AM
I couldnt help but imagine this:


https://www.youtube.com/watch?v=6UfCv1XCMHg

No you don't quite get the KISS approach I mean.? . . . . .My KISS would have been to throw it out the window.!:triumphant: