[Mellbreak]

Good to hear from you again kb0nly but sorry to see you have had bother with your computer. The circuit and board layout diagrams are below...


Attachment 263528 Attachment 263530

The circuit board was milled (of course) and I could provide g-code files if you wish.

The HEX file is attached. A word about the software if I may. It's functional, but should be treated as a beta release. The original source code is written in the excellent Proton BASIC language. If you want the source let me know. It's fairly simple and should easily translate into other BASICS or the likes of C.

A word or two about the software if I may. About 10 times a second the software measures the period that a single PWM cycle is at logic zero and produces a number typically between 6 and 245. At low RPMs this number is high, and at high RPMs low. The number is then averaged over 4 PWM cycles to reduce the chance of ‘flickering’ between consecutive readings.

The next stage is to take the PWM number and convert it in such a way that it can be used to control a digital rheostat. In effect the analogue pot on the front panel is being replaced by a digital one. This is not a simple task. For starters, the spindle does not start to rotate until the resistance has been reduced from about 5k to approximately 3.8k. Consequently an offset must be applied to the number fed to the rheostat chip. The second problem is that whilst the numbers produced by the Mach3 software are beautifully linear, the behaviour of the spindle motor is most definitely not. This means that the software not only has to take care of the offset problem, but it must scale the PWM number to correct for the non-linear nature of the Spindle motor before it is transmitted to the rheostat. This it does well, however there are some interesting issues that make life difficult.

The software aims to spin the motor at a speed (in RPM) near that indicated in the Mach3 software. It is never likely to be an exact match. This is because the spindle speed varies with AC mains voltage in a significant way. The DC supply is not stabilised and there is nothing that can be done about this unless a different, stabilised, power supply is used. A second problem is that the controller has no feedback mechanism which allows the actual spindle speed to be monitored. This means that the software has no method of correcting for motor loading. I’ve tried to compensate for this by having the spindle rotate at a faster rate off load. So, for example, if Mach3 is set at 8000 RPM, then it is likely that the off load spindle speed will be in excess of 9000 RPM. Again the actual speed will depend on the AC mains voltage at the time. All of this said, I’m pretty certain that these problems exist on all the PWM controllers at the budget end of the market.

If this wasn’t enough, the JP-1482 controller has a couple of interesting features. As I have mentioned in a previous post, the speed at which the spindle rotates is not just dependent on the position of the control knob on the front panel. It can also depend on the previous position of the knob! Try marking a position on the spindle control knob scale about 2/3 of the way along. Start from the minimum setting and rotate the knob to the marked position. Note the spindle speed. Next rotate the knob until you get maximum speed and then move it back to your marked point. Measure the speed once more. You'll find there is a several thousand RPM difference! To overcome this problem, the software looks to see if the user wants to increase or decrease the spindle speed. If the spindle speed is being decreased, then the resistance of the rheostat is set to maximum for a period of 25ms before returning to its correct setting. This makes the JP-1482 controller ‘think’ that the control knob is being rotated from the minimum position. Normally the user will not notice this.

The second issue is that, although it appears as if rotating the spindle control knob gives a smooth increase or decrease in speed, I’m not certain that this is the case. I suspect that changes in speed, especially at lower RPMs, are incremental. The practical consequence of this is that occasionally, with small speed adjustments, changing the speed in MACH3 does not produce a change in spindle speed. The software is changing the resistance of the rheostat, of that I’m certain. Further investigation is needed.

MACH3 SETTINGS

The software is built around some specific settings and are the ones recommended in the controller documentation. These are:

Kernel Speed: 35000Hz (Do not change.)
PWM frequency: 200Hz (Do not change.)
Minimum PWM 10%
Maximum pulley speed: 11000 rpm (Don’t set a minimum.)

Please note that the software has a minimum speed of around 2000 RPM.

I’ll probably spend some time developing this board and software for myself. If I manage to make significant improvements, I’ll post them here. If anyone comes up with a better method of controlling the JP-1482 I hope they’ll do likewise.

In the mean time if anyone else does try to build this, I’d be interesting in hearing how you get along.

As always, make sure you know what you are doing around mains voltages. All modifications to your controller are entirely at your own risk.



Bob

[Mellbreak]

Post removed by Mellbreak.

[Mellbreak]

New information added to post 25.

Bob

[haydensmart]

kbOnly,

I have found another way of turning the Spindle on and off without hacking the JP-382A at all.

This is obviously if you have the low voltage issue, and do not intend on using the A Axis.

I connected 5V pin from A Axis to Pin 4 of the Speed controller (4th from the left when looking from the front), then The PU (Pulse) pin from A Axis to Pin 3 of the Speed controller. I then placed a 1kOhm resistor between the two wires.

I intend going an alternate route with my PWM speed control. I tried using the A Axis Direction Pin directly in place of the Potentiometer. As the Pot simply pulls a MCU pin to ground to speed up the Spindle it is limited to 5 V. Using A Axis output you have direct access to the Parallel port pins. Keep in mind there is no optical isolation. The Parallel port can not pull enough current to ground to slow the spindle to a crawl and only gives you speed control from around 70%-100%.

I am going to build a small PWM switching signal through a Opto and Filtering (Maybe RC filter) that will essentially replace the POT.

What are your thoughts.

Great job on your mods, just trying to develop a simpler method.

[kb0cyl]

Ok, i got tired of having to flip the Spindle switch all the time and decided it was time to get Mach3 doing its job!

I have a newer version of the Chinese CNC 3020T-DJ with a black controller box and the controller board is model JP-382A and the spindle board is JP-1482.

The controller has a PWM output and the spindle board has a PWM input but they are not connected. Here is some pictures i included on another post..


After i got done adding home switches to the controller i also started to trace out the boards to figure out the Spindle control.

I just received my 3020T-DJ with the same controller box and powered it on for the first time. It looks like Mach3 is working correctly and I will be modifying the spindle on/off like you have described. Can you fill me in on adding the home switches? More specifically, how to do it and what do I need to add them? I plan on using this for engraving small plastic parts and wood. (kb0nly, it seems we may have gotten licensed the same year. I had a vanity license for a while and then went back to my original call sign)

[kb0nly]

I just received my 3020T-DJ with the same controller box and powered it on for the first time. It looks like Mach3 is working correctly and I will be modifying the spindle on/off like you have described. Can you fill me in on adding the home switches? More specifically, how to do it and what do I need to add them? I plan on using this for engraving small plastic parts and wood. (kb0nly, it seems we may have gotten licensed the same year. I had a vanity license for a while and then went back to my original call sign)

I just saw posts to this thread in my email again while cleaning out my inbox, so sorry for the extremely long delay in getting back here, its been a crazy year with all the work i have taken on in my little shop.

The limit switches are just basic micro switches that i added with some brackets i made up, nothing fancy. I can get some pictures of it if you wanted to see it. The way i set them up in Mach3 though is to use them as homing switches, So i can home all axis before a project. As for what you would need to add them, its just a few basic things, JST connector as mentioned back a page in this thread, then i used the same style microphone plug and panel mount jack to mount it up in the back panel. The wiring was what took the longest as i took the cable chains apart and routed the switch wiring with the existing wiring to make it all neat.

I will have to get some pictures as it would far better explain what i did for them. Unless your worried about crashing it to a limit i wouldn't honestly bother with it, i thought it would be nice to have the homing switches but in the end i generally just zero the axis to the work i clamped down and hardly use the homing feature anymore. But then thats because 99% of the work i do on this machine is making PCB's, and i make an acrylic table thats bolted to the aluminum table with insets routed out for different size PCB blanks. So i just toss in a blank tape the corners down and then zero the engraving bit to the corner and away we go.

[zj4041]

Hi All,

Received my 3040 CNC with 800W VFD Motor few days ago. I Got X,Y,Z communication with Mach 3 software to work.

Bad New: Mach 3 software is not able to turn on/off the spindle or control it. Or even give me a read out on the RPM.
So far I can only control it manually from the inverter. The inverter gives me a Frequency Digital Display (Hz readout though that I can convert to RPM)

Is there an easy fix to this (pin- port config) with out having to make hardware changes?

My controller should be TB6560, but logo on controller says JD.. or HNE something something, see picture below

Attachment 284520

Thank you,
Rob

[AndersL]

You need two JST-XH connectors to make the interconnect cable, also if you want to add limit/home switches its the same 4 pin JST-XH connector, look on eBay for LiPo balancing cables and connectors as they are used in the LiPo battery packs.

Note... I don't yet have the PWM speed control working, still finishing up how to set that up, but i DO have Spindle On/Off control now from Mach3! The PWM control just wasn't as big of a concern for me as i generally run it full speed all the time for PCB milling.

Ok, on the controller board the jack marked PWM i numbered them from the back panel forward, pins 1-4.

1. LPT Port Pin #17 to Spindle Controller PWM Input
2. +5v from Spindle Controller to the EL817 labeled DDO, This is Spindle On/Off
3. Ground from Spindle Controller
4. Spindle On/Off

On the Spindle Controller board i labeled the pin numbers Left to Right looking at it from the front panel, pins 1-4.

1. PWM
2. +5v
3. Ground
4. Spindle On/Off

The front panel Spindle switch only switches a ground on the Spindle Controller and its in parallel with an Optocoupler, part number EL817, on the Spindle Controller board. So you can leave the front panel switch for manual control and leave it in the OFF position to allow Mach3 to turn it on and off.

You need to make a cable so that its matching 1-1, 2-2, 3-3, 4-4, using the pinouts as i described them. The plug orientation is flipped, so pin 1 on plug on one end of the cable would be the opposite side of the plug on the other end. You will understand what i mean when you look at the jacks, the controller board has its jack going front to back with the keyway to the left and pin 1 towards the back, but the spindle board has its jack going left to right and keyway to the back and pin one on the left. So the cable is a flipped crossover config.

Set Mach3 for Active Low on Pin #1 to turn the Spindle On/Off.
Set Mach3 to control the Spindle PWM using pin #17

To sum it up. Is it enough to connect the PWM connectors and then you can drive spindle on / off?

[toppytop]

Don't know if anyone still following this but I have been doing a lot of reading of everything I can find on the web concerning controlling the spindle speed with software with the JP-1482A board. From what I have read this board using a micro controller for the PWM signal is not programmed for external software driving. I looked over the schematics from the PW3618 and the PW3024 boards. People say that they have made these boards operate by software by just connecting the PWM in and out between the two boards in the controller. By my feeble knowledge it looks to me like you could build the 555 timer portion of the circuit used on the 3618 or the 3024 board. Its just an opto isolator, a few resistors and caps and the 555 chip. Plug it into the PWM out from the JP-382A board and feed your new board into the gate of the mosfet driving the spindle motor using whatever method you like. Hopefully someone is still thinking about this also and has a comment.

[SouthPark]

Don't know if anyone still following this but I have been doing a lot of reading of everything I can find on the web concerning controlling the spindle speed with software with the JP-1482A board. From what I have read this board using a micro controller for the PWM signal is not programmed for external software driving. I looked over the schematics from the PW3618 and the PW3024 boards. People say that they have made these boards operate by software by just connecting the PWM in and out between the two boards in the controller. By my feeble knowledge it looks to me like you could build the 555 timer portion of the circuit used on the 3618 or the 3024 board. Its just an opto isolator, a few resistors and caps and the 555 chip. Plug it into the PWM out from the JP-382A board and feed your new board into the gate of the mosfet driving the spindle motor using whatever method you like. Hopefully someone is still thinking about this also and has a comment.

That's the main problem. There appears to be descriptions of ideas or methods associated with modifications to achieve spindle control via software (eg. Mach 3 control). But the described methods never appear to be complete or clear. Eg. " connecting the PWM in and out between the two boards in the controller" isn't very clear. Also, "build the 555 timer portion". What 555 timer? What's that used for?

At the moment, I'm trying to achieve the same thing..... spindle control using software (such as Mach 3), instead of always needing to turn the dial on the black coloured China CNC controller box. I don't know much about the boards in that controller box right now, since I'm new to this. However, I assume that when we turn the physical speed control dial clockwise, this action will vary a DC voltage (somewhere inside the China CNC controller box) for which to rotate the spindle, right? Or, that varied DC voltage is used to generate a PWM signal (with adequate power handling) to energise the spindle. So, if we are able to disconnect that particular internal DC or PWM signal (which means removing the physical dial speed-control functionality), and put in our own hardware module as a middle-man, then we could then have a way of controlling the spindle speed via software.

It appears that people have successfully achieved software-controlled spindle operation, but they all have their own way - due to the various kinds of software and hardware available. What I'm interested in is .....a method to make minimal modifications to the stock China CNC black-coloured controller box in order to achieve spindle control (via software - preferably Mach 3).

This person at the following link appears to have done some very nice work in figuring out various details of China CNC controllers.

https://hackaday.io/project/6776/logs

But...... once again... some inconsistencies in details. Eg.... it mentions that the China CNC spindle dial involves a resistive divider, consisting of 6K resistance, with one end of that 6K resistance connected to "3.3 V".....while their hand-drawn circuit diagram indicates that end is connected to 5 Volt DC, instead of '3.3V'. And then it says the mid-point of the resistive divider has a 5K rheostat (in the form of a potentiometer) connected to it, and the other end of the rheostat goes to ground. However, the details indicate that the 'maximum resistance' of the rheostat is 3.3K, which doesn't make sense when they're initially indicating a 5K rheostat. Confusing.

[SouthPark]

This person at the following link appears to have done some very nice work in figuring out various details of China CNC controllers.

https://hackaday.io/project/6776/logs

But...... once again... some inconsistencies in details. Eg.... it mentions that the China CNC spindle dial involves a resistive divider, consisting of 6K resistance, with one end of that 6K resistance connected to "3.3 V".....while their hand-drawn circuit diagram indicates that end is connected to 5 Volt DC, instead of '3.3V'. And then it says the mid-point of the resistive divider has a 5K rheostat (in the form of a potentiometer) connected to it, and the other end of the rheostat goes to ground. However, the details indicate that the 'maximum resistance' of the rheostat is 3.3K, which doesn't make sense when they're initially indicating a 5K rheostat. Confusing.

Update: After some tinkering of my 3040Z-DQ, I found out (from internet searches) that the Mach 3 software can indeed be used to produce a PWM signal signal from a parallel port pin of a desktop computer. I used windows xp (as Mach 3 recommends XP or Win 2000, which are 32 bit operating systems). Apparently Mach 3 has issues with controlling the parallel port when 64 bit operating systems are used. Need to go to 32 bit systems. I followed a youtube video for making pin 17 of the parallel port produce PWM.

https://www.youtube.com/watch?v=P3CNtyOVHNc

After checking things out, I discovered that the most important things include:

1) Mach 3 configuration of the spindle PWM output pin is performed in the 'motor output' section, in the configuration row for the 'Spindle'. And this PWM output pin will relate to the "STEP" pin. Yes, relates to the "STEP" pin. Yes, we know that "PWM" has got absolutely nothing to do with stepper motor "steps" (since spindle motors are usually not stepper motors), but these are the 'rules' that the Mach 3 software writers created themselves, however ridiculous and confusing it is. Anyway, in the 'Spindle' row settings, a tick/checkmark is needed to ENABLE the spindle, and the "STEP" pin number (which is the parallel port pin number we want to produce the PWM signal) is to be set to 17.

So the 'Spindle' row settings should be:

TICK - 17 - 0 - CROSS - CROSS - 1 - 0. And, in the Spindle Setup page, under 'Motor Control', put a tick/check mark for the "Use spindle control" box, and also put a tick/check mark for the "PWM control" box, and also set PWMbasefrequency to "250"

The settings for output#1, output#2 etc are unnecessary for PWM generation. So, no need to deal with output#1, output#2 etc. I'm just focusing on the essentials for getting a PWM signal to show up on a parallel port pin.

2) Then, massively important, is to finally go to the Mach 3 front control panel, and focus on the SPINDLE SPEED control panel, which is the box that has the software BUTTON that says 'Spindle CW F5". Here, it's necessary to use the mouse to push this button in order to activate it. Activated mode will mean yellow edges on this button. And secondly (massively important), is to click on the purple box to manually set a Spindle Speed. For example, the purple box might initially have '0'. So, it's necessary to use the mouse to click on that value (such as click on the '0' with the purple box around it), and then manually type in something like a value of '100' (then hit enter). Once we've manually entered a value, the PWM pin we selected (such as pin 17) will spring to life. And we can then use the mouse to increase or decrease the PWM duty cycle with those software up/down arrow buttons directly underneath the "Spindle CW F5" button.

I'm writing the above, since I noticed that leaving out some very tiny thing can put us in a 'so close, but yet so far' situation in terms of trying to achieve something that is meant to be simple to do.

Also, for the spindle "PULLEY" configuration, I just configured Pulley Number 1 with Min Speed = 0, Max Speed = 100, ratio = 1.

Anyway, what I need to do now is to just make use of the PWM signal. At least the software allows us to generate PWM.

Now, how could we use the PWM? I'm thinking.....the mid-point of the voltage divider (of the CHINA CNC controller's physical potentiometer) is currently used in the China CNC controller for providing a DC control voltage. So, instead of using the potentiometer for producing the DC control voltage, it should (instead) be possible to make our own control DC voltage by using an "R-C filtered" version of the parallel port's PWM signal. From multimeter measurements on the PWM pin of the parallel port, I found that 100% duty cycle, which translates to a plain DC voltage, corresponds to 3.3 Volt.

[Parkournerd]

Hey kb0nly,

I don't know if you're still on here, but I had questions about the PWM-OUT port. Are you suggesting that the pins are as follows?

1 - PWM Output -> connected to the collector of ENO optocoupler. Anode pin for this optocoupler is cut on some boards (not mine). Connected to the A-axis.
2 - External voltage -> emitter for DDO?
3 - GND -> collector for DDO?
4 - Spindle ON/OFF output -> Q for DDO?

It seems like with the mod, both the physical switch and the software switch must be on at the same time right? Also, it seems like PWM Output could not be enabled without interfering with the A-axis. In this case, would it be possible to output a PWM signal on pins 2-3 instead? That way, the spindle would have to be turned on using the physical button, but then the speed could be controlled by software and *effectively* turn off by decreasing the speed to 0.

Thanks!

[Parkournerd]

I found that you can actually use the pin 4 for speed control as well. But the problem is that for speed control, the duty cycle is kept at 50% and the frequency is varied; at the maximum speed, the frequency is 1.475kHz and the speed is proportional to the frequency, but the problem is that with 0rpm, the frequency goes to 1.475kHz again. Other than never using the maximum speed, how could you overcome this? Is there something I'm missing?

[Maverick17]

Hello
@ kb0nly
Excuse me would you be able to have both schemes in larger 1024x960?
top small to read the diagrams.
cordially

[xx1180]

Ok, i got tired of having to flip the Spindle switch all the time and decided it was time to get Mach3 doing its job!

I have a newer version of the Chinese CNC 3020T-DJ with a black controller box and the controller board is model JP-382A and the spindle board is JP-1482.

The controller has a PWM output and the spindle board has a PWM input but they are not connected. Here is some pictures i included on another post..

Attachment 250510

After i got done adding home switches to the controller i also started to trace out the boards to figure out the Spindle control.

You need two JST-XH connectors to make the interconnect cable, also if you want to add limit/home switches its the same 4 pin JST-XH connector, look on eBay for LiPo balancing cables and connectors as they are used in the LiPo battery packs.

Note... I don't yet have the PWM speed control working, still finishing up how to set that up, but i DO have Spindle On/Off control now from Mach3! The PWM control just wasn't as big of a concern for me as i generally run it full speed all the time for PCB milling.

Ok, on the controller board the jack marked PWM i numbered them from the back panel forward, pins 1-4.

1. LPT Port Pin #17 to Spindle Controller PWM Input
2. +5v from Spindle Controller to the EL817 labeled DDO, This is Spindle On/Off
3. Ground from Spindle Controller
4. Spindle On/Off

On the Spindle Controller board i labeled the pin numbers Left to Right looking at it from the front panel, pins 1-4.

1. PWM
2. +5v
3. Ground
4. Spindle On/Off

The front panel Spindle switch only switches a ground on the Spindle Controller and its in parallel with an Optocoupler, part number EL817, on the Spindle Controller board. So you can leave the front panel switch for manual control and leave it in the OFF position to allow Mach3 to turn it on and off.

You need to make a cable so that its matching 1-1, 2-2, 3-3, 4-4, using the pinouts as i described them. The plug orientation is flipped, so pin 1 on plug on one end of the cable would be the opposite side of the plug on the other end. You will understand what i mean when you look at the jacks, the controller board has its jack going front to back with the keyway to the left and pin 1 towards the back, but the spindle board has its jack going left to right and keyway to the back and pin one on the left. So the cable is a flipped crossover config.

Set Mach3 for Active Low on Pin #1 to turn the Spindle On/Off.
Set Mach3 to control the Spindle PWM using pin #17

Now, as i said i haven't hooked up and configured the PWM control to pin 17 yet, so all i wired up was three pins between the boards, i didn't install the wire from pin #1 on the controller to pin #1 on the spindle controller, thats the PWM control. You need pins 2-4 for On/Off control as the Spindle controller board feeds +5v back to the controller board via pin #2 to run the Optocoupler labeled DDO, the output of which toggles another EL817 optocoupler on the Spindle board that then enables the spindle motor, the front panel switch is wired in parallel across the Optocoupler on the Spindle controller board which gives manual control and can be left in place.

Here is some schematics i drew up quickly... Notice that pins i didn't continue to trace out for PWM on the Spindle board or the ENO Opto on the Controller board are marked unknown. Also in the case of the EL817 marked ENO on the Controller board note that Pin 1 of the component was clipped and bent back at the factory, ENO looks to have something to do with the A-Axis as its connected to one of the pins of the four pin header marked A-Axis, i didn't continue tracing that out as it wasn't necessary. Also pin 2 of ENO dead ends at a set of open pads on the back side of the board, i would guess there should be a 330 Ohm resistor there as they kept the same basic design across the boards, using a 4.7k pullup resistor to 5v on pin 1 and a 330 ohm resistor on pin 2 on the others for what i would assume is current limiting through the optocoupler.

Attachment 250418 Attachment 250420

Anyway, with a three pin jumper you can make the Spindle On/Off control work with Mach3, or whatever software you use, but i haven't got the PWM config working just yet. I would assume it requires removing the front panel trimpot from the circuit, either just unplugging it or installing a jumper as in the other versions of these boards but i don't have that info yet.

do you have a photo with the cable in place between the two boards

[Cellfreak]

Thank you for finding the spindle control.

[pache11]

This is an old thread, but I hope this will be helpful.
The controller supports a display and switches to change mode. This Hackaday post shows how to connect display and change controller modes.
I have used this successfully and modified the display board to use it's buttons to change modes.
After changing mode to PWM the 4 pin connector works for on/off/pwm
https://hackaday.io/project/6776-304...g-machine-mods

[JayBee44]

The controller supports a display and switches to change mode. This Hackaday post shows how to connect display and change controller modes.
I have used this successfully and modified the display board to use it's buttons to change modes.
After changing mode to PWM the 4 pin connector works for on/off/pwm

I was messing with things this weekend and managed to use that post (and a couple others) to successfully get spindle on/off working via software and get my estop so it actually functions on my Chinese 3040 CNC. Good to see you've got PWM working. That hackaday posts mentioned one of the issues is he can't get 100% spindle speed in PWM mode. Instead he achieved only about 63% of full speed. Have you noticed this? I currently use Mach 3 and was wondering if I'll encounter this issue. I'm happy with at least finally having basic on/off spindle control via Mach 3, but PWM speed control would be nice too!