[daniellyall]

I think that calls for a teck to come and answer, I don't touch AC I have a qualified person for that stuff, but one thing I do know and have seen is high and low voltage wires need to be all by there self not running side by side, side by side is a big problem.

the DC should have a common there will be but if one of them is wrong it can loop the noise through everything what it sounds like it is doing from the drivers, it's close to what I had but that was a bad AC earth leak and one shield connected at each end what was very hard to find it was years.

[johnno402002]

I've managed to get the noise on the spindle drive signal down a long way. I noticed that two of the three axis drives were causing the noise, the middle and the top one. The way the wiring was tied together both of these had a fair length of their motor cables running beside their ac supply input, where the bottom quiet drive did not . I separated the wiring, and I put a couple of turns of each motor wiring through a ferrite toroid, the same as the factory did on the spindle motor cable. There is still a small amount of noise, but its way better than it was. I can now get the spindle down to 120 or so revs, and the phantom typing into the keyboard at low revs has gone as well. So, one issue down.

The spindle changing speed by itself has not happened again since we found the loose connection on the feedback cable from the motor, so that is now ok as well.

The speed is still out a bit, but now that I have the map of the differences I can compensate for it.

The only remaining question is about the settings in the spindle drive for differential in PID. The setting instructions above from Defeng have the proportional and integral, but do not mention the differential. My understanding is the differential is the key setting for how the spindle reacts to a change in load, so it would be handy to know if it is adjustable.


[johnno402002]

Picture of new wiring layout at the axis drives

[daniellyall]

that's better

[Skyfire]

Hi John,

I heard about your thread from a friend and I have not logged in the forum for a long time. As you guys can imagine, I'm drowned in the works of many subjects to improve my machines. I had planned many times to update something here and gathered a lot of materials but just got no time. I reviewed your problems and then write an email to you but you might didn't receive it. We do provide post sale services but I missed your emails to my former temporary email box. Anything please just email to our @skyfirecnc.com email boxes then I or my workmates will see and response.

Anyway, after reviewing your problems, I thought them through and here might some correct answers:

1. about the noise. As you have successfully located the noise source---2 of the stepper drivers---I have 2 most possible reason here

a. We tested the machine for a long time before shipping and I have not seen this noise problem. I think the most possible reason is that we have different power supply voltage. In China, our power voltage normally between 210v-220VAC. But I heard in AU, it will be about 230-240VAC. Then after the transformer, the stepper driver will get about +10% voltage. And this may make the drivers work at the unstable status and then generate some noise and transmit the noise to the controller via power supply circuit.

b. another possible reason is that I simply ordered fake LEADSHINE drivers... and the fake ones don't work stable when the power voltage is higher than in China. I had managed to make the steppers work at a higher end of it's permitted voltage range to gain higher torque and speed. This might not good for your higher local power voltage.

Power voltage difference is the major problem I known from the feedback of initial users. My solution is to add a 240VAC input pin for the transformer. Or switch power supply.

2. about analogue output linear. The JIAMEN motion card has this nature shortcoming. I hope I can explain this clearly. actually, Jiamen card analogue output is not a real analogue signal. It actually outputs the PWM signal but at a high base frequence(1kHz? I can't remember clearly now) The PWM signal will then get "integral" to simulate analogue output. It's a cheap solution and not accurate. I had suggested Jiamen to provide pure analogue output or just real PWM with base frequency 10-100Hz BUT THEY JUST DON'T DO IT..Actually my BLDC driver supports PWM input and even can use RS485 speed control. They can have less than 0.5% speed control tolerance easily. But with Jiamen card, PWM is not an option.

Under this situation, I decided to turn to LINUX CNC or other higher end PC based controllers, or industrial CNC controllers next on my higher end machines. Jiamen card will just be used on lower end solutions. Not only because it doesn't have good spindle speed control features, MACH3 is now not satisfied when we hoping for encoder support, AC servos, ATC, rigid tapping, REALTIME control...etc.

3. about my BLDC driver and motor. Frankly, my BLDC controller is not the best, but can say it reliable after many version developments. And all its parts are real deal for VFD. It's PI control but not PID because it uses AVR IC but not a DSP. AVR is not fast enough to do PID. But PI is still enough for BLDC motor control as an economic solution. My next plan is to update the BLDC motor with encoder and then use STM32 on driver for closed loop control. We can expect the BLDC system will be much more powerful ---especially at low speed end. Current HALL feedback can NOT provide enough PPR to tell the driver how to do at low speed end. Especially when speed control signal is not stable---the BLDC driver will be crazy...

Then the BLDC motor.. It's fantastic but also a big hole for electric engineering. It's much more complex than AC(servo) motors. I suffered a lot to find the best PI for it. Good news is that our current BLDC motor is much much better, and then it's much easier to get it under control. I can't explain too much here because it's a big story to talk.

Then my solution on this subject is: for lower end machines, I have the developed BLDC system. for higher end machines, AC SERVO. I even developed our own AC servo spindle motor and just about to start using it now.

So anyway, I would like to do better and better. Many apologize to the unsatisfied issues accrued. I'm working hard to make machines better and provide better, faster services. I will post my many progresses soon.

Thank you to all of you guys!

Defeng

[ihavenofish]

so, on the bldc... what would it take to get full performance at all speeds, but cost relatively little and have it quickly (not years)?

i ask cause i have 2 of these, and many other people have them lying around, plus of course skyfire owners.

[handlewanker]

Hi Defeng.....as long as the various boards that get updated are marked with some form of coding or part number to easily and quickly identify them when a user, (not all that tech savvy), or anyone acquiring a machine with these electronic parts has a problem, and needs to know if the problem is due to an old board or obsoleted design that is not totally compatible with current motor usage etc.

It's good to have updated and top of the range equipment for a machine when you buy it, but later, when more updates are available or a complete revision of the system(s) is current, the buyer needs to have access to the old stuff if that is only suitable with their particular design build......maybe years later even etc.

Changing a driver....whatever..... to improve the working capability for the latest set-up is fine, but if it needs a more updated motor to also improve the capability, that can be expensive to replace the whole packet.

We are totally in agreement that performance with cutting edge technology is paramount to be competitive in an advancing CNC market, but reliability is even more important for previous buyers who are using the same technology that was developed years earlier.

Unfortunately the computer is also a prime suspect in this equation when it comes to glitches in a system that just won't go away, and as computers go out of fashion with their design and electronic types, that is just another problem to throw money at.

How long Mach3 will remain the mainstay of the control market only the gods know....and Microsoft is becoming less god like by the year with every model rendition.........I don't even want to think about something that also needs to be updated and then won't work with anything you have.

I think if a discount policy was in place for original equipment buyers to enable them to acquire the latest design electronics when all else fails, then this would soften the blow when the bullet has to be bitten.....LOL.....sometimes adding patches to an old coat is not the way to go and this is also evident with later CNC developments and design changes that call for a complete revision rather than a fixit.
Ian..

[Skyfire]

so, on the bldc... what would it take to get full performance at all speeds, but cost relatively little and have it quickly (not years)?

i ask cause i have 2 of these, and many other people have them lying around, plus of course skyfire owners.

I have to tell you that it pretty much depends on the motor itself. BLDC motor is much complex in design and manufacture than AC motors. Many small tricks in its production affect its last performance. Unfortunately we don't have many choices. Many suppliers have <1.2 kw BLDC motors and they are ok. But up to 2.2kw, there are not. Earlier SIEG had a sub company can provide good BLDC motors but they can't reach high speed; Sanmutian as you know, is almost the only choice. Their former 122mm frame BLDC motors are not quite stable and generate the BLDC performance problems. I dis assembled all of the sample motors from suppliers and I know what makes the motors not satisfied. I even thought about building the BLDC motors with some friends but it's a huge invest and can't promise the time to make a good motor. Good news is that their current 130mm frame new BLDC motors are pretty good! much much more stable with my driver---Even with my oldest driver version. I approved this and using the 130mm motors now.

So if you let me know what brand/type BLDC motors you have, email me and I might be able to tell you if they are improvable. Like stepper motors, if your motors had experienced high temperature, crash or some other damages, it's inner magnet field might turned weak. then the motor is totally not savable.

As to my machine owners, we always keep the upgrade services at cost price. And promise free replacement in guarantee period for the failure parts.

[Skyfire]

Hi Defeng.....as long as the various boards that get updated are marked with some form of coding or part number to easily and quickly identify them when a user, (not all that tech savvy), or anyone acquiring a machine with these electronic parts has a problem, and needs to know if the problem is due to an old board or obsoleted design that is not totally compatible with current motor usage..
Ian..

Hi Ian,

Yes keep the long term support and upgrades even to the earliest version is important. I have all versions of designs documented and spare parts available, So I can provide long term support even after years. I might suggest changing some parts like motor or driver, but no intention to earn any more money from my machine owners but just want to improve their machine performance. So as I said only cost price to upgrade the necessary parts.

I can frankly tell you guys that, I didn't sell many machines in last 2 years (so I have no much trouble of upgrading issues?) because I want to make the machines as good as I can so keep developing stuff costed most of my time even missed many business chances. Until now I really feel much good and have strong confidence with my machines.

Everything has its limitation. for example the BLDC motor, we might not happy with its performance under some conditions, but frankly, except it, we can't have any better choice to gain such torque output at such compact size and cost. lol

Anyway, don't worry about upgrading support. It's not a big issue to us.

Thank you

Defeng

[johnno402002]

Hi John,

I heard about your thread from a friend and I have not logged in the forum for a long time. As you guys can imagine, I'm drowned in the works of many subjects to improve my machines. I had planned many times to update something here and gathered a lot of materials but just got no time. I reviewed your problems and then write an email to you but you might didn't receive it. We do provide post sale services but I missed your emails to my former temporary email box. Anything please just email to our @skyfirecnc.com email boxes then I or my workmates will see and response.

Anyway, after reviewing your problems, I thought them through and here might some correct answers:

1. about the noise. As you have successfully located the noise source---2 of the stepper drivers---I have 2 most possible reason here

a. We tested the machine for a long time before shipping and I have not seen this noise problem. I think the most possible reason is that we have different power supply voltage. In China, our power voltage normally between 210v-220VAC. But I heard in AU, it will be about 230-240VAC. Then after the transformer, the stepper driver will get about +10% voltage. And this may make the drivers work at the unstable status and then generate some noise and transmit the noise to the controller via power supply circuit.

b. another possible reason is that I simply ordered fake LEADSHINE drivers... and the fake ones don't work stable when the power voltage is higher than in China. I had managed to make the steppers work at a higher end of it's permitted voltage range to gain higher torque and speed. This might not good for your higher local power voltage.

Power voltage difference is the major problem I known from the feedback of initial users. My solution is to add a 240VAC input pin for the transformer. Or switch power supply.

Defeng

Hi Defeng,

Thank you for responding, I do appreciate it. I did get your email and responded to it, but it must not have got through.

I think you are correct with the possible cause of the noise being the voltage difference, I have checked the voltage in my shed, it gets over 250 volts at times. The spindle is running much better now, quite usable as it is.

I had a look inside one of the leadshine drivers, it looked very nice, but I guess if it was a fake I would not be able to tell the difference. The axis drives do seem to work very well.

The non linear spindle speed is ok now that I know it is like that, I can compensate for it to some degree.

Thanks for confirming that the driver is PI and not PID, I was wondering if it was just setting or not, so I wonder no more.

I am very happy with the machine now, it is incredible value for the money it owes me.

Thanks again Defeng

John

[Skyfire]

Hi Defeng,

Thank you for responding, I do appreciate it. I did get your email and responded to it, but it must not have got through.

I think you are correct with the possible cause of the noise being the voltage difference, I have checked the voltage in my shed, it gets over 250 volts at times. The spindle is running much better now, quite usable as it is.

I had a look inside one of the leadshine drivers, it looked very nice, but I guess if it was a fake I would not be able to tell the difference. The axis drives do seem to work very well.

The non linear spindle speed is ok now that I know it is like that, I can compensate for it to some degree.

Thanks for confirming that the driver is PI and not PID, I was wondering if it was just setting or not, so I wonder no more.

I am very happy with the machine now, it is incredible value for the money it owes me.

Thanks again Defeng

John

Thank you John for your prompt feedback! I have been keeping eye on this issue since I knowing your problem. And It's such a relief to hear your SVM-2 works normal now.

Top 250VAC is too high for your SVM-2 as its standard requirement is 220VAC. I even feel lucky nothing burned at so high voltage! LOL But this had reminded me to consider the wider power input range for my machines. So it did helped me to make the machines better. I have made closed loop steppers as my bottom configuration recently. It will alarm if the input voltage is too high.

Any more support you need just email to [email protected] or [email protected]. or add Skype directly: ID skyfirecnc

We will help on your issues asap.

Thank you John

Defeng