help! X3 spindle problem

[presslab]

I too had problems with my old electronics, shorting on the scale. Here is a list of components I had to replace:

Main board:

Designator Package Replacement part
7806       TO-220  ST L7806
Q1         SOT-23  MMBT2907
Q2         SOT-23  MMBT2222
Q3         SOT-23  MMBT2907
Q4         TO-92   PN2222

Motor:

Quantity 3 hall effect sensors. Old part number was UTC U18, replaced with
Allegro A3280.

http://www.utc-ic.com/spec/U18.pdf

Front panel board:

Designator Package Replacement part
R22        1/8W TH 47 ohm
U3         SO-14   ST LM324
U2         DIP-8   X9C103P
U7         DIP-4   Fairchild H11A817C
U8         DIP-4   Fairchild H11A817C

I since have received the upgraded boards. The circuit interface is quite different between the front panel and the main board, it is some kind of synchronous serial interface and it seems like it is polling the pushbuttons.

In my journey to upgrade to CNC I have come up with a plan. The Syil spindle board is not the best because it does not have fast ramp times and I'm unsure how good semi-rigid tapping will work with that. I don't really want to use the old main board as it is pretty dodgy. I occasionally had to whack my e-stop relay to get the machine to work and it does not have isolated inputs.

My plan is to use an aftermarket brushless amplifier for the spindle with analog inputs. The old front panel controller should work with this and I can also easily interface it to Mach3. With the addition of a simple relay I can select from the front panel or Mach3 control.

[neilw20]

I have been designing BLDC circuitry upto 2KW so I am comfortable renovating the existing unreliable,cobbled together lot.
I have used the the MC33035, MC33039 controller setup and they work really well when implemented correctly in proper current mode with slope compensation, and proper PID design of the speed/power control loop.
You can even dynamically brake the MC33035, which goes towards solid tapping....

I am going to come up with a fix on the whole thing, including some new small control boards. The interface system lost the plot when it went CNC.
Before that it still sucked.

[presslab]

That sounds like a fun project. I'd like to see what you come up with.

I too have designed some BLDC servo controllers for work, around 500W, with CANbus and s-curve velocity profile generation. Right now I think I can find a used brushless amplifier on ebay for $100 or so, and that's my plan.

With one of those amps, a SSR for estop, a relay for switching from manual/auto control, and a few discretes I think I will have something.


Another thing worth noting is the spindle motor. The "old electronics" motor has a voltage rating of 110V, and the "new" one is 220V. The RPM listed is the same, 3500 RPM. The old amp must have driven the motor differently or something.

I think I will switch back to the old motor, and with a aftermarket amp I should be able to get some more spindle speed. This is better than a pulley change because I will still have low speed torque.

[ddexd]

As far as spindle speed goes, I did notice when I was making repairs to the control boards when I bypassed the digital potentiometer that I could get the spindle speed up to 2500+ rpm. With a working digital potentiometer installed the maximum spindle speed is only 1730 rpm.

Neil, definitely would like to see your board design to replace the standard SX3 control boards.

[presslab]

Originally Posted by ddexd As far as spindle speed goes, I did notice when I was making repairs to the control boards when I bypassed the digital potentiometer that I could get the spindle speed up to 2500+ rpm. With a working digital potentiometer installed the maximum spindle speed is only 1730 rpm. Neil, definitely would like to see your board design to replace the standard SX3 control boards.

Oh good, I'm glad to see that it can go faster. Thanks for that. I did try the old motor with the new circuit boards and it did not accelerate correctly. It sort of jumped a bit on acceleration which I'm guessing would be incorrect feed forward / PID tuning in the new controller circuit. It didn't go any faster, either; I think the controller is closing the velocity loop.

[adel_acameron]

Hi Chrisjh, can you please provide details of the switching chip IC please. My board failed only after 2 hours of use (I hope it hasn't blown the main motor controller as well). The problem is that my board was wedged in an it got damaged as I was dismantling it so I now have a broken board which I am wonder in I can repair. So anyone who can provide a circuit diagram I would be grateful, or any advise on where I can get a replacement.


Thanks

Alex

[neilw20]

If it is a late model Motor control PCB in the X3 the switchmode chip (u201) is VIPer22a from ST
Data:
http://www.google.com.au/url?sa=t&ct...kUBsRSFbsF7e8w
If it is an early PCB without the piggy back PCB is a TNY268PN
Data sheet:
http://www.google.com.au/url?sa=t&ct...6NeYNSxXot1gvw
I know most of the chips that have been ground off.
IMHO the older PCB is a better design.
Chinese grown BLDC can't be as good a tried and tested design from Motorola.

They don't grind the tops off to hide them from us here.
It's so their CN neighbors don't clone them!

Neil

[adel_acameron]

Thanks for these pictures they were able to help me identify a small surface mount resister than just "brushed off" the board (so much for the soldering techniques) when I was pulling out my failed controller. I saw it tumble into a sea of saw dust and filings so I have no hope of recovering it. So can anyone help and tell me what value R23 is on the controller board please - also any help with part numbers would help as well.


Thanks

Alex

Originally Posted by ddexd neilw20, thanks for the help. Just in case the hall sensors are out I went ahead and ordered a few from Digikey. By the way the Allegro A3141 hall sensors have been replaced with an updated model Allegro A1101. I traced S1 down and it goes to the tapping switch. Connections K3 & K4 are wired to the ESTOP button and the chip guard door. The two switches appear to be wired in series. The small pink colored connector is wired to the control panel mounted on the front of the mill. I've attached a couple of photos of the connector and the pin labels. The first 3-photos are of the control panel and the last is of the motor PCB. I also put in a call to Grizzly tech support and I was promised that one of their electrical support folks would be calling me back within 24hours.

[neilw20]

Alex.

Can you take a picture of your problem area.

I haven't got access to back of SX3 at present, but will have a look if your PCB matches my model PCB's.

Are you taking the 7 hour drive to the BBQ?

Neil