[James Newton]

Ah... I understand. The 3 axis full kit order button simply populates a shopping cart with the items needed. It will put in 3 Linistepper kits. Simple change the quantity to 4 and press enter. The shopping cart will apply the same discount to the 4th. It will also suggest other optional items that you might want like the parts for the relay driver.

[Dirty Steve]

Ok, new problem today on a blue 4-axis tb6560 board.

my 4th axis is crapping out, when I jog the axis forward it turns, but multiple forward jogs can spin motor both ways, same for reverse jog.

Had the motors tuned and set in Mach3 and this happened during a code run.

If I plug the motor to another axis it works fine, so I am thinking the problem is in my 4th axis driver.

Could this be a bad capacitor on the driver or the tb6560 chip itself?

Motor goes forward and back randomly when holding a jog button.

[doorknob]

Ok, new problem today on a blue 4-axis tb6560 board.

my 4th axis is crapping out, when I jog the axis forward it turns, but multiple forward jogs can spin motor both ways, same for reverse jog.

Had the motors tuned and set in Mach3 and this happened during a code run.

If I plug the motor to another axis it works fine, so I am thinking the problem is in my 4th axis driver.

Could this be a bad capacitor on the driver or the tb6560 chip itself?

Motor goes forward and back randomly when holding a jog button.

Do you have access to an oscilloscope (or to a friend who has an oscilloscope)?

It would be worthwhile monitoring the signal on the DIR pin of the TB6560 from the failing axis (possibly triggering on the STEP pulse to that axis) to see if there may be a problem with the timing or shape of that signal.

Have you bypassed the optoisolators on that board, or have you done any of the other modifications to that axis? I know that you talked about jumpering the optos on a board in a previous post, but I'm not certain whether this is the same board. If it is the same board, did you solder the jumpers or put wires in sockets (that may have come loose)?

[Dirty Steve]

no access to a scope.

Was all working fine.

The only mod i have done is to jumper the optoisolators.

Motor will reverse direction back and forth randomly when holding a jog button.

Other axises are working without issue.

[doorknob]

no access to a scope.

Was all working fine.

The only mod i have done is to jumper the optoisolators.

Motor will reverse direction back and forth randomly when holding a jog button.

Other axises are working without issue.

Without using a scope you are pretty much left to guesswork.

Maybe try unplugging and replugging your cable from the PC, and checking or rejumpering the optoisolators for that axis (soldering, if not already done).

I suppose that it could also be due to a poor solder joint somewhere on the board, or an errant signal from the pendant electronics - does the problem still occur if the pendant is unhooked? (I am guessing that you are jogging the axis via Mach3 and not by the pendant, because you saw the error during a code run, but maybe my guess is wrong - please clarify).

[Dirty Steve]

never got the pendant to work, jogging in Mach3.

Unplugged and reconnected everything, same problem.

[doorknob]

While it is not out of the question that you have a faulty TB6560 chip, it is probably more likely that something outside of the chip is amiss - for example, even a hairline crack in a circuit board trace could produce strange symptoms.

So locating a friend with an oscilloscope may yet be the best next move.

[Dirty Steve]

have swapped out capacitors, swapped out the two 74ac14 chips, no change.

checked traces, checked solder joints, can't see any breaks.

Ordered a couple of TB6560 chips, worth a shot at $5 a chip.

Motor operates correctly if plugged into any other axis.

So damn close to 3d printing!

[spiderdab]

I don't know mach, but if there is the possibility you could try to invert signals on parport of motors 3 and 4, for example. if the problem stay on the 4th axis, it's for sure un the driver, while if the problem swap on the 3rd axis it could be a software fault.. then i would also try to change the parallel cable.

[H500]

Most likely problem is a blown 6560. It might be wise to disconnect the motor from that channel, just in case the current limit no longer works. A burn motor is even less fun.

[Dirty Steve]

TB6560 chips on order. Motor runs fine on another axis. Have checked everything I can at this point, waiting on new chip. I think it's the chip because everything was working fine, then crapped out in the middle of a code run.

Here is a video clip. Correct steps/per setting is around 67, if i crank that up to 1000 steps, the motor jogs, but randomly reverses during jog in Mach3. Excuse the low quality video.

MOV04031 - YouTube

[Dirty Steve]

It was a bad TB6560 chip. Replacement arrived in the mail today, installed and problem is fixed!

[skyguynca]

First of all thanks to everyone who submitted mods and all the help.

I do think this board is worth buying and the mod is cheap, and well I put it thru hell today and it still works somewhat, well part way.

Well I did the mod that is in the "chinese board fix updated" but messed it up a bit and blew out the mod 74h14 chip, see pic below.

Did another one and installed it right this time so it got 5v instead of 24v and it works!!!

However only the Z axis works now, y just makes the motor whine and x does nothing at all.

I will try replacing the TB6560 chips first and see it that brings it back online unless you guys think of something else to check first. I do not have a osciliscope. Just a meter

Thanks for help

[aarggh]

Seeing as you've reworked the board, the first thing to do is maybe check the traces carefully to see if there's a solder blob cusing a short or a lifted/broken track. Then start looking at chips.

cheers,
Ian

[skyguynca]

Done that, no shorts or blobs.

I followed a bad picture of the trace to solder the ground to on the mod board, well it got hit with 24v instead of 5v and popped the chip. I am sure that I did some damage to the driver board just not sure what because I can not see any physical damage.

[tufane]

I think these chinese boards (3-4-5 Axis) are comletely wrong designed. Boards are not competible to TB6560AHQ datasheet page 28 "Power-on Sequence with Control Input Signals". All TB6560 inputs have internal pulldown resistor and inputs must be LOW state during power up. But these boards supplying 5 volts to all inputs and not reseting the chip.

Datasheet says "If this sequence is not properly followed, the IC may not operate correctly, or the IC and the peripheral parts may be damaged".

I have 3 Axis board and I'll remove all pull-up resistors, add delay circuit for VMA and VMB, and add few 4066 switch ICs for delayed power-up switching for all inputs. But it will take time....

Regards.

[James Newton]

tufane, I totally agree with your assesment. But think about how much money and time you will have invested in making that 6560 board work reliably! Instead, you could upgrade to the TB6564AHQ which is /NOT/ sensitive to power up sequence. It's currently available as a kit, but you are going to be doing surface mout re-work, so I don't think soldering together a (mostly) thru-hole kit should scare you:
THB6064AH 4AMP Stepper Motor Driver Kit. massmind
AND it supports 4 Amps at 50 volts... and it /actually/ supports that.

[skyguynca]

I built a modified momus with lead screws instead of belts, with the modified TB6560 board, I got the directions from here. I finished it and everything seemed to be working fine until I noticed something tonight.

I wrote a g code program tonight, the main goal when it is finished is to make a cpu liquid cooling block.

So I wrote the first part that goes and drills 5 rows of .125 holes .250 deep.

Later I will add to the code that will connect each of the holes on both the x and y axis with a .0625 channel so all that will be left is a bunch of pins for the water to flow around to remove heat from the CPU.

Well anyways I drew out the pattern on a piece of 3/8 MDF and showing each location that should be drilled. I wrote the code so that it stops just above the board and does not drill it, didn't want to make a mess and keep the garage quiet while writing the code and testing it.

I wrote the code two ways, one will just drill the holes at a slow feed, the other will peck drill to depth.

The program ran great but I noticed when peck drilling by the 5th row the bit was actually touching the board. So I re-homed the machine and started the program again. Well it happened again. So homed again this time I measured the final depth on every position. It was loosing 1mm per 5 pecks in each of the 9 holes in 5 rows with 45 holes total so 225 pecks.

So I tried the slow drill program with out the pecking, same think occurred but it took running the program 5 times. So that equates to the same number of changes in direction to achieve the same loss.

Hmmm, I checked for binding, all felt smooth and good. Turned down the feed rate, tuned the stepper for slower accel and velocity, turned down as much as 25%, made no difference.

Anyone have some suggestions of where to check next?

[skyguynca]

Wow, that is it, that fixes the problem, thanks so much for your so fact based insight. You are sooooo smart, who knew you had all that problem solving expertise............certainly no one here knew it.

It's rumoured that everytime someone buys a TB6560 based board, an engineer cries!TB6560 based board, an engineer cries!

[Philipf]

Could anyone advise me on the best way to bypass 6N137 and EL817 optos? I have a dead driver and believe all signals are stopping at 1 of the 3 optos.

Thanks alot, Phil