There is no solders because I'm using an header. I don't want to have to solder all the 25 pins each time I'm doing a test
Just to see if the issue was coming from the header, I just tried on a brand new board without the header, and I just burned another TB6560. I putted a sink and a fan on it just to be safer, and was able to burn it even at 12V
Only 2 TB6560 remaining. After that I will move to a Gecko drive...
I was able to get some results few minutes ago with another TB6560. But 2 outputs seems to always be low. I will have to figure that. There might be some bad connection somewhere. Ouf. Long road!
Are you able to do any sort of diagnostics on the circuit (multimeter, oscilloscope, etc) before it burns up, or does it burn up too fast to even watch what is happening?
One problem with using a header to connect the chip is that the header will not be very good at sinking the heat that builds up in the chip. The usage considerations suggest using a heat sink, too.
The usage considerations also suggest The power-on sequence described on page 28 must be properly followed But I haven't been able to find the page they are talking about. Perhaps there is some trick to powering up this chip???
Do you happen to have a fuse between the power supply and the chip like the usage considerations indicate? I've been wondering if a fuse could prevent a total burn-out like this.
Also, did you ever get some non-inductive resistors? If I remember correctly, there was some confusion about that a couple weeks ago.
The pin matrix for "Possibility to smoke or burst" on page 12 of this document does indeed seem a bit comical, if not downright knee-slapping worthy of a chortle.
One problem with using a header to connect the chip is that the header will not be very good at sinking the heat that builds up in the chip.
Here is what I have now fixed in the back of the TB6560: http://tiny.cc/x0xg0 But I did not used that at the beginning, and that also, I think, was the cause of some burns.
Do you happen to have a fuse between the power supply and the chip like the usage considerations indicate?
Yep, I have. But since my power suply is made for 3A, I have 3A fuses. anyway I just bought a real power suply from DigiKey (http://tiny.cc/kyrd2) so I will be able to limite the current now.
that would be if it was accompanied by jmspaggi's photograph.
Good idea! I might have to send them an email with the suggestion
@all: Thanks for your suggestions, comments, help, etc. Today I will order 4 more TB6560, a real power supply and some missing comment. And that will be my last try. so by the end of the coming week-end, I might be able to give you more news.
I burned few other TB6560, and I still don't know why....
JM,
on a different thread, someone posted the following product PCB that seems to use the same chip but to control 3 axes. If you have a look, it's got a fan and some serious heat sinks, so perhaps one of your problems is that you just need a lot of heat sinking capability when working with this chip???
on a different thread, someone posted the following product PCB that seems to use the same chip but to control 3 axes. If you have a look, it's got a fan and some serious heat sinks, so perhaps one of your problems is that you just need a lot of heat sinking capability when working with this chip???
I was not using heat sinks before, but now I'm using huge one Each TB6560 now have a sink as big as the one in the URL you gave, and each with a fan. http://tiny.cc/x0xg0
Okay I now have more faith in you design. Phil over at Pminmo has a pretty good reputation for designing drivers that work. Did you buy the board from him?
Not directly from him. I bought that on eBay. Phil did a USB driver also, with a guy in Canada, and I bought the card from that guy. He is the one selling and sending the cards in North America.
I will bring him all the components today and he will solder everything for me. And Sunday, I will get everything back. So the big final result will be Sunday evening... I will cross my fingers all over the week-end
They was all working fine with 12V, 1A. Motors was moving.
But as soon as I tried with higher values, I burned them. One by one.
So I gave up with the TB6560 and bought a Gecko 540 yesterday that I received today.
And my motors are already moving with 24V, 2A with no issue. I will try later this afternoon with the full power supply (36V, 2.8A per motor). But I'm confident.
The conclusion is: The TB6560 is probably is good component. 1/16 steps, easy to use, etc. But I'm not good enought in electronique to use it. The card ElectricAye sent the link might be a good alternative, but I choosed to buy a made in noth-america card instead
Now, I have to find how to pulse properly with the propeller to do some testing.
First I tried it without the motor because it's far from my propeller. When I saw that 10 turns will be done in 0.94s, I was pretty septic, thinking it's should not be working.
But when I tried with the motor I was impressed!
I tried it many times, and look like the motor is stopping exactly at the same position. I can't garanted it's not missing a step, but look like it's accurate.
I can't also garanteed it's doing 10 turns because it's going to fast
I will try to run it for 100 or 1000 turns tomorrow to see if it's running well over time.
When you said "settings 'above 12,000 will cause the G251 to falter." is it because the propeller will not be fast enought to handle that?
Will it be faster with PASM?
Also, what's the voltage you are using for your motor?
Not a high enough voltage for the motor coils (unlikely)
Improper ramping (possible)
I also think the limitation is coming from the spin code. My voltage is 30V. I will have 40V at the end, so I will try this evening with the max power supply, but since the Propeller is takin 0.94 to pulse, the motor can't turn faster.
I think so.
***Please note that if this driver is rewritten in PASM, please post the PASM code in this thread so that others may benefit from it. Also try to increase the 12,000 setting, and please add notes of your test results pertaining to accuracy and speed.
The PASM code will not come from me It will take me months to do it. There is code in you SPIN that I don't even understand. So I will try, but don't expect any result before a couple of weeks. But it will be very interesting to have it.
Thanks again for the code. I will look at the PASM version starting next week-end. Maybe we will have something before next summer
Comments
Just to see if the issue was coming from the header, I just tried on a brand new board without the header, and I just burned another TB6560. I putted a sink and a fan on it just to be safer, and was able to burn it even at 12V
Only 2 TB6560 remaining. After that I will move to a Gecko drive...
JM
Maybe that can help YOU.
Look attached PDF
I already have this one too.
I was able to get some results few minutes ago with another TB6560. But 2 outputs seems to always be low. I will have to figure that. There might be some bad connection somewhere. Ouf. Long road!
JM
Sorry to hear this.
Are you able to do any sort of diagnostics on the circuit (multimeter, oscilloscope, etc) before it burns up, or does it burn up too fast to even watch what is happening?
One problem with using a header to connect the chip is that the header will not be very good at sinking the heat that builds up in the chip. The usage considerations suggest using a heat sink, too.
The usage considerations also suggest The power-on sequence described on page 28 must be properly followed But I haven't been able to find the page they are talking about. Perhaps there is some trick to powering up this chip???
Do you happen to have a fuse between the power supply and the chip like the usage considerations indicate? I've been wondering if a fuse could prevent a total burn-out like this.
Also, did you ever get some non-inductive resistors? If I remember correctly, there was some confusion about that a couple weeks ago.
http://www.semicon.toshiba.co.jp/docs/application_note/en/TB6560AHQ-E_en_090324.pdf
The pin matrix for "Possibility to smoke or burst" on page 12 of this document does indeed seem a bit comical, if not downright knee-slapping worthy of a chortle.
And you know what? I think that's why I burned the few first chips. The sequence is described on page 1 in the document Sapieha posted above.
Here is what I have now fixed in the back of the TB6560: http://tiny.cc/x0xg0 But I did not used that at the beginning, and that also, I think, was the cause of some burns.
Yep, I have. But since my power suply is made for 3A, I have 3A fuses. anyway I just bought a real power suply from DigiKey (http://tiny.cc/kyrd2) so I will be able to limite the current now.
I fixed that too. I used http://tiny.cc/in7ii instead. Seems to be better.
@all: Thanks for your suggestions, comments, help, etc. Today I will order 4 more TB6560, a real power supply and some missing comment. And that will be my last try. so by the end of the coming week-end, I might be able to give you more news.
JM
JM,
on a different thread, someone posted the following product PCB that seems to use the same chip but to control 3 axes. If you have a look, it's got a fan and some serious heat sinks, so perhaps one of your problems is that you just need a lot of heat sinking capability when working with this chip???
http://www.aliexpress.com/product-gs/340635688-3-axis-TB6560-3-5A-CNC-engraving-machine-stepper-motor-driver-board-16-segments-stepper-motor-wholesalers.html
I was not using heat sinks before, but now I'm using huge one
Not directly from him. I bought that on eBay. Phil did a USB driver also, with a guy in Canada, and I bought the card from that guy. He is the one selling and sending the cards in North America.
I will bring him all the components today and he will solder everything for me. And Sunday, I will get everything back. So the big final result will be Sunday evening... I will cross my fingers all over the week-end
JM
They was all working fine with 12V, 1A. Motors was moving.
But as soon as I tried with higher values, I burned them. One by one.
So I gave up with the TB6560 and bought a Gecko 540 yesterday that I received today.
And my motors are already moving with 24V, 2A with no issue. I will try later this afternoon with the full power supply (36V, 2.8A per motor). But I'm confident.
The conclusion is: The TB6560 is probably is good component. 1/16 steps, easy to use, etc. But I'm not good enought in electronique to use it. The card ElectricAye sent the link might be a good alternative, but I choosed to buy a made in noth-america card instead
Now, I have to find how to pulse properly with the propeller to do some testing.
JM
I will have to move to the mechanical part now that motors are moving.
JM
I just tried you code.
First I tried it without the motor because it's far from my propeller. When I saw that 10 turns will be done in 0.94s, I was pretty septic, thinking it's should not be working.
But when I tried with the motor I was impressed!
I tried it many times, and look like the motor is stopping exactly at the same position. I can't garanted it's not missing a step, but look like it's accurate.
I can't also garanteed it's doing 10 turns because it's going to fast
I will try to run it for 100 or 1000 turns tomorrow to see if it's running well over time.
When you said "settings 'above 12,000 will cause the G251 to falter." is it because the propeller will not be fast enought to handle that?
Will it be faster with PASM?
Also, what's the voltage you are using for your motor?
Damn, I'm impressed! I like what I saw
JM
I also think the limitation is coming from the spin code. My voltage is 30V. I will have 40V at the end, so I will try this evening with the max power supply, but since the Propeller is takin 0.94 to pulse, the motor can't turn faster.
The PASM code will not come from me
Thanks again for the code. I will look at the PASM version starting next week-end. Maybe we will have something before next summer
JM
Also, here is a tentative of PASM conversion.
I'm trying to translate that:
repeat RampingSteps phsa := HIGH_PULSE_WIDTH waitcnt(Counter += CycleOffset--) StepCount++Into that:
MOV _index, _RampingSteps loop1 MOV phsa, HIGH_PULSE_WIDTH ADD _Counter, _CycleOffset WAITCNT _Counter SUB _CycleOffset, #1 SUB _Index, #1 TEST _Index, #0 IF_A JMP :loop1On the SPIN side of the PASM version I also have this initialisation:
And those new variables:
I think the loop can be improved. I remember I saw a way to do -1 on the test. That will remove the SUB line. But I did not find it.
Anyway, I have not yet tested it since I don't have a propeller at the office.
JM
PS: I just figured that "MOV phsa, HIGH_PULSE_WIDTH " is not compiling....
In the meantime, here is my quick and very dirty PASM code I'm trying to build:
PUB PulseTheStepPinWithCog(StepPin, TotalSteps) | RampingSteps, RunningSteps, Counter, CycleOffset if TotalSteps > RAMPING_RANGE RampingSteps := SPEED_RANGE RunningSteps := TotalSteps - RAMPING_RANGE else RampingSteps := TotalSteps / 2 RunningSteps := TotalSteps // 2 CycleOffset := MIN_SPEED ctra[30..26] := %00100 ' Configure Counter A to NCO ctra[5..0] := StepPin frqa := 1 dira[StepPin]~~ Counter := cnt LONG[_RampingSteps] := RampingSteps LONG[_RunningSteps] := RunningSteps LONG[_CycleOffset] := CycleOffset LONG[_Counter] := Counter COGNEW(@cogstart, @cogstart) DAT ORG 0 cogstart MOV _index, _RampingSteps loop1 MOV phsa, _HIGH_PULSE_WIDTH WAITCNT _Counter, _CycleOffset SUB _CycleOffset, #1 SUB _Index, #1 TEST _Index, #0 IF_A JMP loop1 MOV _index, RunningSteps loop2 MOV phsa, _HIGH_PULSE_WIDTH WAITCNT _Counter, _CycleOffset SUB _Index, #1 TEST _Index, #0 IF_A JMP loop2 MOV _index, _RampingSteps loop3 MOV phsa, _HIGH_PULSE_WIDTH WAITCNT _Counter, _CycleOffset ADD _CycleOffset, #1 SUB _Index, #1 TEST _Index, #0 IF_A JMP loop3 end JMP end _index LONG $00000000 _RampingSteps LONG $00000000 _RunningSteps LONG $00000000 _CycleOffset LONG $00000000 _Counter LONG $00000000 _HIGH_PULSE_WIDTH LONG $FFFFFFF6It's compiling, but I don't garanteed the result
anyway, I'm working on that.
JM