[RedskinsJBS]

The way I understood it, the 'semi-log' adjustment was chosen because it provides a better feel to the user. It seems like this is a firmware setting, which may mean that this could be as easy to change as a different version of the firmware. This, of course, is all hearsay by yours truly.

If this is the case, then one could take a few voltage measurements from the pot and see what router rpm they correspond to. Then one could calculate a curve based on those measurements and develop a conversion table to be used in Mach when controlling the speed. Even if Mach is not able to do this conversion automatically, how difficult is it to look up 1 value on a table?

I'm thinking that a resolution of 250 RPM would be more than enough in the table. And a very accurate curve could be developed with just a few measurements, so a measurement every 1000RPM would be way more than plenty. 1 every 5000RPM would most likely give an accurate-enough RPM control. Plug everything in to a spreadsheet and viola!

I'll make the table if someone is willing to take the measurements. Otherwise, you will all simply have to wait until I get my Super-PID


I did, order my own 'modified' version of the Super-PID. I plan on using it with the Porter Cable 7518 router in my CNC build, so I needed the added heft of the modified version. Is it odd that I ordered the router speed controller before I have a machine to put it on?

Jay

[RomanLini]

CarveOne- Good to hear you received the high-current SuperPID all ok in the airmail.

Khalid and Al_the_man- Thanks guys for the hardware suggestions re direct connection to Mach3 via the PC. I have some options already under consideration for future revision of the Super-PID.

RedskinsJBS- Thanks for your order! Also you are right about the semi-log system used for the Super-PID speek knob, more info below.

-------------------------------------------------

Re Mach3 controlling the RPM;

I'll try to break this down to show where it's at at the moment.

1. The Super-PID was designed so it can be operated from a 0v-5v analog voltage to control the RPM. This is standard, similar to many VFDs etc.

2. If you already have a 0v-5v analog control voltage it can be connected instead of the Pot Output wire "PO" and will control the speed. However this raises 2 issues;
A. The 0v-5v range will produce 5000-30000 RPM result.
B. The 5000-30000 is semi-log, not exactly linear.

Both A and B were done deliberately to give the very best possible "feel" of the Super-PID speed knob. Believe it or not my very first pot code was linear; 5000-30000 with 17500 centre of the dial. But it sucked, the low RPMs (that are so critical for tuning cutting) were very touchy on the dial and the 20000-30000 RPM range took up half the dial. Imagine turning the knob 54 degrees at the bottom giving a 100% RPM rise, but the same 54 degrees near the top of the dial gives only a 20% RPM rise! That's the trouble with "linear by RPM" it is terribly clumsy on a speed dial.

So I deliberately coded the pot "feel" to give the very best usability, with a graded semi-log effect so turning the knob 5 degrees near the bottom gives a similar perceived RPM % change as turning the knob 5 degrees near the top. I even expanded the range a little around 10000 RPM giving it even better tunability there. This was tweaked over weeks of me testing and doing cutting jobs etc.

So, for Super-PIDs already in service; you can use a 0v-5v analog speed control voltage but it will give 5000-30000 RPM range and semi-log (so centre dial is about 11000 RPM).

-------------------------------------------------

Hardware options to get 0v-5v from Mach3;

1. Connect the PWM output from Mach3 through any CMOS logic chip (like Khalid's circuit in post #192) then through a 47k resistor and 100uF cap to filter the PWM. Total cost under $2 but is fiddly and you need electronics skills.

2. If you already have the Mach3 PWM signal from a CMOS IC (not direct from the parallel port) but from a buffer IC on a break out board, you only need the 47k resistor and 100uF cap. Many people's break out boards are already set up like this, with CMOS buffers.

3. Or use a third party module to convert the Mach3 PWM to analog 0v-5v signal. Ger21 already posted a link to two products that do this, in post #38. These may or may not need the additional 47k and 100uF filter parts.

4. Gecko G540 users can use a couple of 22k resistors to turn its 0v-10v output into 0v-5v. Then you still connect a 47k resistor and 100uF cap.

-------------------------------------------------

Firmware solutions for now.

I can offer a re-programming service for existing Super-PID users, to make your speed control go from 0-30000 linearly. Note that under 5000 RPM will still be fixed to a 5000 RPM minimum, but the 0-30000 RPM will correspond with Mach3. Please remember this will make your speed control knob feel awful, so this option is best if you want to use Mach3 for speed control only.

I'll discuss with Val the possibility of offering this firmware as an option with new SuperPID purchases. It may add a couple of days to your order time.

-------------------------------------------------

Future product Mach3 compatibility.

For the next PCB revision it looks likely we will add some extra hardware to allow direct PWM input into the Super-PID. This may add a few dollars to the price.

I can also add a switch option to allow changing between 0-30000 linear speed and 5000-30000 semi-log speed, so the user can use either system as desired.

The next PCB version will be some months away, depending on sales. I'm sure Val will post any new product updates well in advance.

Sorry for the long post!

[RedskinsJBS]

Sorry for the long post!

I can live with long posts if they thoroughly answer all of the questions asked! Thank you very much for your time.

Now, I am going to assume that I could wire up a SPDT (SPCO) switch in line with the 'sense' terminal on the POT and I would be able to switch from POT speed control to Mach3 speed control?

Is there any need to wire the grounds to a switch as well?

I think initially all of my speed control will be done by hand, but once I get comfortable with this whole thing, I may move on to allowing Mach to control the router speed.

[lancut]

"Gecko G540 users can use a couple of 22k resistors to turn its 0v-10v output into 0v-5v. Then you still connect a 47k resistor and 100uF cap."

I'm getting closer to pace an order. Can you post some sort of illustration/ schematic how the capacitor and the resistors should be connected? Thanks.

"Sorry for the long post!" No apology necessary, after all, it benefit those who want/need to know about the product...

[ger21]

Hardware options to get 0v-5v from Mach3

1. Connect the PWM output from Mach3 through any CMOS logic chip (like Khalid's circuit in post #192) then through a 47k resistor and 100uF cap to filter the PWM. Total cost under $2 but is fiddly and you need electronics skills.

Roman, would it be possible to draw up a schematic for us electronically challenged?? The circuit Khalid posted had both a forward and reverse, and was a bit unclear to me.
Since I'll have a free wire in my SuperPID case, I would like to give this a try. Still haven't placed my Mouser order, so I'll kill a whole flock of birds with one shot.




Off to the garage for more wiring fun.....

[revwarguy]

Thanks for the clear explanation, Roman. Makes perfect sense.

I guess I'll try the 22k voltage divider with the 47k series resist and the 100 uf cap across the G540 output. We'll see how that goes. If that's not good, can I just order a replacement firmware as opposed to sending one back? When I get this working, I have the feeling I'm not gonna want any downtime with it!

Looking forward to playing with this - still waiting for my order! Its a long way from there, eh?

[ger21]

still waiting for my order! Its a long way from there, eh?

Mine took about 2 weeks to get here.

Here's today's progress. One hour of staring at my control box, and one hour to connect 2 wires. It's a bit hard to get a good pic of the display, as it over exposes.

Need to mount the router, solder the sensor, and connect the 110V. May take until Friday night.

[lancut]

It looks good! Waiting for Fridays report.

[CarveOne]

Mine took about 2 weeks to get here.

Here's today's progress. One hour of staring at my control box, and one hour to connect 2 wires. It's a bit hard to get a good pic of the display, as it over exposes.

Need to mount the router, solder the sensor, and connect the 110V. May take until Friday night.

Gerry, That's a good looking heat sink add-on. I thought about adding a plate sort of like that but had not thought about adding the holes. For the PC7518 Super-PID I found an old PC PSU circuit board that has a heat sink on it that I can steal. Another idea I had is to add a scrap piece of aluminum angle or aluminum plywood edging and bandsaw some kerfs into the "fin(s)".

CarveOne

[RomanLini]

RedskinsJBS- You are correct, if you want so switch between the speed knob and Mach3 input you just need a SPDT switch, to changeover that one wire.

Thanks also re your suggestion before of charting the Super-PID semi-log knob. I'll try to get some time to do that over the next day or so.

Lancut- Thanks for the suggestion, I will make up a simple wiring diagram of how to connect that.

Ger21- You want a schematic of how to connect a CMOS buffer IC? Sure I will draw one up too. Remember if your PWM signal has already come from a buffer IC on a breakout board there is no need for a separate IC.

Good to see your Super-PID light up! Those LCD displays are hard to photograph. I look forward to seeing yours cutting too! Check those mains wires are screwed down tight.

Revwarguy- You're an electronics guy so you should have no trouble wiring up those 3 resistors and a cap. That will get it up and running fine with 5000-30000 control. Val has approved me handling chip exchanges, although the concept was as a turnaround chip exchange not chip sales.

[ger21]

Ger21- You want a schematic of how to connect a CMOS buffer IC? Sure I will draw one up too. Remember if your PWM signal has already come from a buffer IC on a breakout board there is no need for a separate IC.

Connected directly to a parallel port pin.

Good to see your Super-PID light up! Those LCD displays are hard to photograph. I look forward to seeing yours cutting too! Check those mains wires are screwed down tight.

I'm going to redo them tonight, as one pair is backwards, since I couldn't see the numbers on the wires when I attached them.

[Khalid]

Just installed the super-PID, wired and tested.. awesome man... At 5000 almost no noise, at 12000 i am hearing it louder but How was i working in intense noise at 25000 for the past 4-years... That was a torture really..

I am searching for my camera to post some pictures and video but still not found... Okay going to search it again..

[Khalid]

Although i was busy today in some other activities but i got some time and completed Super-PID cabling and tested successfully... Attached are the self explanatory pictures..

The Super-PID enclosure look dirty but i will install a colored screen labeling all buttons....

I will post details of all by tomorrow .. Soon i will upload the video..

[CarveOne]

Looks good Khalid! You are going to like your new setup.

CarveOne

[lancut]

"Just installed the super-PID, wired and tested.. awesome man... At 5000 almost no noise, at 12000 i am hearing it louder but How was i working in intense noise at 25000 for the past 4-years... That was a torture really.."

Nice setup Khalid, did you try it under the load meaning making a actual cut, how it is performing?

[Khalid]

Nice setup Khalid, did you try it under the load meaning making a actual cut, how it is performing?

Thanks lancut, Very soon i will test it on actual work piece and show the results.... Currently very excited about the quietness in my CNC room...

My CNC machine looks more scientific ... Its now special :cheers:

[GGagnon]

My SuperPID is model V1.1 no used previously.
When the sensor is pointed on white surface, the graph bar is max and min when pointed on dark part of the shaft.
When the router rotates, although the wave shape 0-5V and a good square wave close to 50-50%, the bar-graph is jumping left and right, the display always shows "Spindle View --r" and no rpm indication.
Any idea what is going on and what can be done next?
Thanks,
Gaston

[GGagnon]

My SuperPID is model V1.1 no used previously.
When the sensor is pointed on white surface, the graph bar is max and min when pointed on dark part of the shaft.
When the router rotates, although the wave shape 0-5V and a good square wave close to 50-50%, the bar-graph is jumping left and right, the display always shows "Spindle View --r" and no rpm indication.
Any idea what is going on and what can be done next?
Thanks,
Gaston

For the record.
The answer to this "testing the SuperPID" problem is:

Condition: 5V applied but no AC
The microcontroller will do its boot up and a static test of the sensor can be perform.
But if you try to make a dynamic test with a rotating spindle fed by an independent source, the bargraph will just be flashing and the RPM indicator will never come on.

Condition: 5V AND Ac applied
The SuperPID will measure speed and show rpm, even with no spindle connected. This applies equally in Open Loop and PID (stable) mode.

[kiwibrick]

Roman,
I have emailed Val and my board is now on the way for repairs.

Thanks,
Damon

[pcneal]

Received my Super PID last week and all is working well so far. My PC892 has been modified for speed sensor (see pict) and temp sensor (see small wire going down side of router with temp sensor mounted near bottom bearing). My SPID has been setup for computer speed control therefore using the linear pot, speed control at the slower rpms is not very sensitive (all covered in earlier posts). I remembered however that you can modify the output curve of a linear pot with the addition of a couple of parallel resistors. My suggestion is shown is the two pictures of the circuit and circuit output. You can move the "logarithmic like" curve around by varying R1 and R2. The circuit's load on the 5V source never falls below 25k (ignoring the input Z of the SPID). This circuit seems to give good control and sensitivity in the low rpm range at the sacrifice of course adjustment at the higher rpms. You could use a "log pot" but they are sometimes hard to find. Just a suggestion Cheers, Parris