[Mauser]

So, the problem is "Inrush Current". Interesting.

It seems to me that if 2.25 A is normal and acceptable, if you could set the limit to 9A (not 10) that should serve the right balance between protection and utility.

Is it possible to stagger the startup of each drive?

[keebler303]

So, the problem is "Inrush Current". Interesting.

It seems to me that if 2.25 A is normal and acceptable, if you could set the limit to 9A (not 10) that should serve the right balance between protection and utility.

Is it possible to stagger the startup of each drive?

Mauser

It sounds like startup is controlled by the drive so staggered startup would require different components for each drive. Best to make them all slower and the same.

Matt

[cnc_4_me]

Mauser

It sounds like startup is controlled by the drive so staggered startup would require different components for each drive. Best to make them all slower and the same.

Matt

If the drives had staggered startup then you would loose coordinated motion.

[cnc_4_me]

Mariss, I am not understanding something here. If the motor drive will see 2.25A at 50vdc then it will see more current at the minimum advertised voltage the G540 can run at, 18vdc. Shouldn’t the current at 18vdc plus a safety margin be the over current set point on the G540.

[dhelfter]

Mariss, Wow I feel a bad for you now :violin: In my for humble opinion I would not trash the 1500 g250's. I would raise the trip level via the simple resistor swap. I do realise that is not being "as safe" however does not some responsiblity of design and wiring need to fall onto your customers? I think it is awesome the way you listen to our imput and fix things for free, that is truly unheard of these days, however you design/manufacture/sell/service only a component of a machine. You should not be responsible for your customers mismatching, or poorly designing their setup. Just my 2 cents (worth less than that!!)
Thanks
Dan

[harryn]

I think you have actually two different business situations with your customers:
- Experienced OEMs
- DIY, one off builders

As well as two different technical situations:
- Applications that will push the 540 to its limits
- Applications that will use perhaps 75 - 80% of the capability

From a practical perspective, the DIYs (such as me) will continue to routinely make wiring mistakes. They will benefit the most from the best protection.

The OEMs are less likely to make wiring mistakes, yet, their feet are held the most to the fire for system performance.

There are also applications where a 3 port 540 might be acceptable vs a 4 axis version. If you doubt this, think of the AMD quad core processors, which had only 2 or 3 cores running. It is a good way to make up for product yield challenges.

This means you could make your product into "3" products:

a) 4 axis, fully protected, reduced total power (more or less your current model, with some more info in the datasheet )
b) 4 axis, semi-protected, full power (proposed model with resistor change)
c) 3 axis, fully protected, full power (current model with one port plugged)

No matter what you say or write, you are going to end up eating the warranty on the "b" version to keep customers happy. Just keep this version feeding the OEM market only, where they are more likely to understand what is going on better.

I know it complicates your product line, but I can see routers using 2 x 540-Cs for 6 motors pretty easily, so the 2 extra ports might not be that important, especially if a person got them for just a few bucks less.

You could try some bench testing of the higher trip level and short some out to see what happens. I am not sure how many would need to be tested in order to actually tell statistics on it.

One of the German firms I do business consulting for goes to great lengths to build high product quality / low warranty cost, and this really does help them. Nonetheless, they bought some computers (for the product) with defective components and it has been time consuming to recover from, although we have worked through it.

I guess my point is that the proactice approach you are taking here will help keep you from experiencing a total loss on those boards and still keep your customers happy.

[LazerBee]

Would a switch mounted on the board going from factory 8 amps to 10 amps be feasible. This way during machine setup the G540 could be in 8 amp mode later allowing the user to select 10 amp mode.

[keebler303]

Would a switch mounted on the board going from factory 8 amps to 10 amps be feasible. This way during machine setup the G540 could be in 8 amp mode later allowing the user to select 10 amp mode.

Not a bad idea at all. :idea:

I thought only windows had a "safe mode"

Matt

[Scramjet]

Good day

As i stated at begining of this thread, (Reply 14), it happens on small motors, i was using only 3 G250 on the G540 (x,y,z) not 4 same time, motors were small 3A on X, 3A o Y and small Vexta .8 A on Z, and it was randomly faulting, i changed to larger 1.4 A on Z and it gone better, but keeps faulting.

I have some other G540 installed in larger machines (4 of 3.2 A steppers) and arent faulting and interchanged same good G540s in small machine and it came fault, and small machine G540 installed in bigg machines and never faulted

So question is: While smaller the stepper, more fault? and bigger steppers no fault?, so small machines need to install bigger steppers?

Thanks for advice, since i need to check my production line.

[keebler303]

I think it has more to do with inductance and resistance than size. The inductance of the motor will affect the current draw at startup.

Matt

[Mauser]

I think it has more to do with inductance and resistance than size. The inductance of the motor will affect the current draw at startup.

Matt

Exactly. Inductance is a sort of resistance caused by the magnetic field in the motor "pushing back" against the current flowing through the windings. But at startup, the "Back EMF" hasn't formed yet, so rather than an inductive load, you've JUST got the resistance of the windings. It's almost like shorting the driver for brief microseconds.

Ironically, the smaller motors might actually have a HIGHER inrush current than the bulkier ones, even though they draw less current once the inductive load is established.

[Mariss Freimanis]

I have opted for raising the over-current protection threshold from 8A to 10A. This is done by changing the value of 1 resistor on the motherboard and it doesn't involve any changes on the 4 drives. This option was chosen because testing showed supply current approached 8A when four 3.5A motors were heavily loaded at high speed. A "soft-start" current control can't get around that condition; only a higher current threshold can.

For the electronics types: This isn't a classic case of inrush current; the phenomena is entirely different.

Imagine you have a loss-less (100% efficient) motor and drive. The motor is stopped and 3.5A is circulating in its winding. The PWM modulation then is 50%; the drive draws 3.5A from the supply and then returns -3.5A to the supply for the same amount of time, 20,000 times a second. The average current from the supply would be zero.

Now imagine you initially power-up the drive and motor. The motor current is at 0A when you do. The PWM duty cycle stays at 100% until the current in the motor rises from 0A to 3.5A, at that point the operating current is reached and the PWM duty cycle drops to 50%.

The rate current change (current slope) can be calculated as V/L; for a 4.8mH motor at 48VDC it's 10,000 Amps/second. Since we are only going to 3.5A, the time to reach this current becomes 3.5A / 10,000A or 350 microseconds. The PWM would be at 100% duty cycle for 7 switching cycles (350 us / 50 us) and then drop to 50% thereafter.

The energy stored in the motor inductance is L * I^2 / 2 or 0.03 Joules (1 Joule = 1 Watt-second). The power required to deliver this energy in 350 microseconds is 84 Watts (Watts = Joules / seconds) which comes out as 1.75A from a 48VDC supply (I = W / V).

1.75A is for an ideal motor and drive. Real-life motors and drives have losses which is why we read 2.25A observed instead of the ideal 1.75A for the inductive energy build-up current; the difference covers these losses, most of which are in the motor. The 24W of losses apportion as 5W in the G250 drive and 19W in the motor. This meshes nicely with theory because the observed supply current drops from 2.25A to 0.5A for a stopped 4.8mH motor after power-up.

Mariss


P.S. Mauser, I think you can see from this the inductive energy build-up current for an ideal 3.5A motor is always the same at 1.75A (motor current / 2). Only the time interval for this current changes; cut inductance in half and the time interval is half as long. Cut the supply voltage by half and the time interval doubles. The current stays the same.

[LazerBee]

So then, what would account for g540's working fine for a period of time on a given setup then faulting several months later. Could this be motor windings breaking down?

[Mariss Freimanis]

Well, there is a purpose for the over-current protection circuit. It's there to protect the G540 and its drives in case of a motor winding short-circuit. If it wasn't there, the first (and last) warning you'd have of a problem is a puff of stinky smoke. Instead you get that irritating red LED.

Generally, if you have a machine that's been working well for a long time and now develops random faults has everything to do with your wiring. The weakest link in any machine is wiring failure at connectors. Months or years of cable flexing takes its toll. Wires flex and ultimately break, almost always at the wire to connector junction.

You need to eyeball every wire at every connector. You may find one or more that have experienced fatigue failure. Electronics aren't like bananas; they can't spoil and deteriorate after a period of time. Either they work perfectly or they don't work at all.

Mariss

[Dylwad]

I want to add too, I originally used the cheap 18 gauge shielded stepper wire available everywhere and it failed in the middle of the wire in a cable chain after a few years, due to wire fatigue. Symptom was random erratic z moves. Really surprised me that the driver didnt smoke, using 251s(same as in the 540)

Replaced all wires with expensive cable tray rated wires so it wouldnt happen again.

[tjb1]

So is this something we can have fixed if we have the problem or should they be sent in now to be fixed/replaced? Mine is only a couple weeks old and I havent got all 4 motors running on it yet so I am not yet sure if I will run into this problem.

[Mauser]

For the electronics types: This isn't a classic case of inrush current; the phenomena is entirely different.

Imagine you have a loss-less (100% efficient) motor and drive. The motor is stopped and 3.5A is circulating in its winding. The PWM modulation then is 50%; the drive draws 3.5A from the supply and then returns -3.5A to the supply for the same amount of time, 20,000 times a second. The average current from the supply would be zero.

Now imagine you initially power-up the drive and motor. The motor current is at 0A when you do. The PWM duty cycle stays at 100% until the current in the motor rises from 0A to 3.5A, at that point the operating current is reached and the PWM duty cycle drops to 50%.

Interesting. A lot of what I know about motors is based on PMDC motors, but the more I learn about steppers, the more wildly different these animals appear to be.

So these drivers know to run 100% duty cycles until the motor is "Charged up" so to speak? What would happen if the drivers simply started at 50% duty cycle?

P.S. Mauser, I think you can see from this the inductive energy build-up current for an ideal 3.5A motor is always the same at 1.75A (motor current / 2). Only the time interval for this current changes; cut inductance in half and the time interval is half as long. Cut the supply voltage by half and the time interval doubles. The current stays the same.

It's still quite a bit to wrap my head around, but it's interesting to learn.

[dhelfter]

Mariss,
Just wondering where we are at on the revised/repaired drives?

Thanks
Dan

[Scramjet]

does this case is over? what was the solution ? . thanks

[Mariss Freimanis]

Mauser,

Step motors are high pole-count PMSM (Permanent Magnet Synchronous Motor) type motors. Their closest cousin is the BLDC (BrushLess DC) motor. There is precious little difference between them except for pole-count (50 poles vs. 6 poles).

You can actually run a BLDC motor as a 3-phase step motor.

Mariss