-
hey everybody
I bought a TB6560 based 4-axis stepper driver year ago from eBay for my DIY CNC router. when that arrived ,i discovered one of the axis has loosing STEPs problem but all other axis were smooth. i changed Motor but still same problem, changed setting to full/half/microsteps but that didnt solve problem and finnaly i changed TB6560AHQ IC but nothing got beter. i leaved that for year but yesterday an accident happened and i lost one of the my smooth axis (TB6560AHQ IC burned) so now i need your help to fix that axis with loosing STEPs problem. i read something about using additional buffer (74LS14) to amplify STEP pulses but couldnt understanding that cause i havnt any experience about electronics.
NOW i wanna ask, could any1 explain me how to fix my board (step by step)?
im waiting for your HELPs.
thanks for reading and sorry about my bad english.
-
1. Replace the 6560
2. Bypass the optos
3. Change the oscillator frequency
4. Remove the current reduction circuit.
The details are posted elsewhere in this thread.
-
I would like to mention that the TB6560 heats up fast and goes into a fail safe mode within a few minutes of starting. Each TB6560 IC needs a fan to keep it cooled. The TB6560 is really a good IC, however, little thought was put into your driver therefore there is little you can do to solve your issue other than replacing the driver with a Gecko. Sorry but It makes absolutely no sense to throw good hard earned money toward a driver that will only fail again in the near future.
[quote=###OCT###;1088244]After reading almost all in this thread, I'm confused.
other than several fixes and workarounds reported by different users,
my version of this board is a little different.
It doesn't have the fan/heat sync combo, no onboard relay, the leds apparently work as supposed, out of the box.
Here is a picture of it (taken from one of the chinese sellers):
[url=http://www.aliexpress.com/product-fm-img/289933517-FreeShipping-Aluminum-box-CNC-3-Axis-Stepper-Motor-Driver-Board-TB6560-Controller-wholesalers.html]Wholesale 3 Axis Controller - Product Picture from John Liao A's store[/url]
More details and manual for this crude contraption, including descriptions about the sections on the board:
[url=http://www.aliexpress.com/product-fm/289933517-FreeShipping-Aluminum-box-CNC-3-Axis-Stepper-Motor-Driver-Board-TB6560-Controller-wholesalers.html]Wholesale FreeShipping! Aluminum box CNC 3 Axis Stepper Motor Driver Board TB6560 Controller[/url]
[IMG]http://img.alibaba.com/img/pb/811/306/363/363306811_117.jpg[/IMG]
[IMG]http://img.alibaba.com/img/pb/980/423/363/363423980_821.jpg[/IMG]
[IMG]http://img.alibaba.com/img/pb/511/264/363/363264511_608.jpg[/IMG]
If someone reading feels inclined to help, but needs more detailed pictures, please let me know and I will provide some.
History, trouble description and fix attempts (I'll try to describe quickly):
0- preamble:
My cnc mechanic parts and functions are fine, as I was able to confirm when my friend came with his drivers , and I was able to machine a few pieces,
using his drivers with my 3 steppers, on my pc. I checked dimensions and finishing on the machined pieces and they were fine.
1- PROBLEM:
Missing, skipping steps on the y axis with my board (see above). Happens every time I try to machine something, very bad. other axis are fine.
2- FIX ATTEMPTS:
software - changed from mach3 to emc2. Successful tests performed with my friend's drivers were done in emc2.
That's pretty much all I've done, other then checking steppers wiring, testing the dip switches (all 3 fine), changing decay and current settings.
edit 03/23/2012: yesterday I replaced the TB6560 IC on the faulty axis, the problem persists.
Running the board at 24v, 10A power supply, 75% current setting (my steppers are 2A/phase), and changing current settings didn't help at all.
If some good soul in here could point where/what to do to fix this, I would be very grateful.
Any help much appreciated.
thanks![/quote]
-
1-This board is not the same design as the one covered in great detail in this thread. The fixes MAY apply to it but you are on your own to figure it out. Following the hints and tips here may cause more damage than good in your case.
2-This board runs on 12 to 24 Volts only as per the silk screen (but not the User guide, 36V).
3-This board uses the aluminium case as a heat sink. Fans not required.
4-If the TB6560 IC chip self destructed, clearly this board has similar flaws as the other TB6560 based designs to it and your best bet is to cut your losses and get bettre than.
5-Looking over the board itself it looks to be of a bettre quality than the others offer with better optos and buffers, separate 12V (DC/DC conv) and 5V sources. Fast diodes for protection. But if it is missing steps, self destructing IC, it clearly suffers from the same TB6560 limitations as the other designs and is a waste of your time and money at any voltage above 12V and over 1A.
You will probably work your way to earning your electronics Eng. degree getting it to work proper and not self destruct every few weeks or months.
MP*10
-
[quote=###OCT###;1088629]Yes, I did that, even with z. I have a secondary stepper motor set, similar specs, 3. tested with them as well.
I believe the problem is in the board, y axis only.[/quote]
Then try this. Map the parralell cable signals to a different motor, do not swap the motor connections. This may tell you if the problem is setup or the board itself.
MP*10
-
1 Attachment(s)
I was going to PM him, but I see that the fella that started this thread hasn't been here in a while. Maybe one of you more intelligent electronics types can answer a question. I read the problems that have been posted and looked at the fixes, but it looks to me that all of the fixes are actually in the breakout section of the controller board. I decided to give the TB6560 a try, but not the integrated version that has the breakout board and 3,4, or 5 driver sections. I will be using a separate breakout board and individual single axis TB6560 drivers like the one in the attached picture. Like I mentioned, it looks to me like all of the fixes are in the breakout section. Is this correct? Does that mean that the single axis boards like I plan to use don't need to be "fixed"??
Thanks in advance.
-
That's where the bulk of the problem is. There might be issues with the oscillator frequency as well as the idle current reduction.
A single axis design could have all the same design faults also. The simple fix is to just bypass the optos on the step and dir signals.
Some are saying that the tb6560 chip itself will self destruct simply from turning the power supply on and off, because it require a special power sequencing, which none of the circuits have. I don't know how serious of a problem this is.
-
I have been tinkering with the 4 chan version of this board for a while now.
I don't think whoever designed the board had much experience designing electronics.
It is littered with 6.8 meg resisters that are basically shorted out by other resistors, either on the board or in the input structure of the 6560 itself.
As has been noted the most, if not all, of the optos can be bypassed, if, for no other reason than because of the way they are powered they provide little in the way of isolation between the parport and the board.
I think some or all the 7414 inverters may be superfluous as well. These and the optos are just causing delays and signal degradation.
-
Hey, thanks guys. I have one of the single axis boards coming to play with. I guess I'll try it out and see if I encounter any problems. If I wind up having to bypass the optocouplers I'm sure I will be back with more questions. I'll keep you guys posted. Thanks again.
-
The TB6560 chip itself will randomly fry if the power up sequence isn't correct. Please check this thread: [URL="http://www.cnczone.com/forums/stepper_motors_drives/139306-how_can_any_tb6560_driver.html"]http://www.cnczone.com/forums/stepper_motors_drives/139306-how_can_any_tb6560_driver.html[/URL] for more information and possible fixes.
And keep in mind that it isn't as powerful as people think. It supports 3.5 amps PEAK, which means for a second or less, but every damn seller lists the driver as supporting 3.5 or even 4 amps, which is a bald face lie. And worse than that, when you look at the actual power dissipation, (page 6 on the data sheet)
[url]http://www.toshiba-components.com/motorcontrol/pdfs/TB6560AHQ_AFG_E_2003_20080407.pdf[/url]
it handles LESS than 43 watts (43 watts with an infinitely good heat-sink, which doesn't exist) which is only a bit over 1 amp at the max 40 volts. Even with a 24 volt supply, it will fry above 43/24=1.8 amps. Only using 19 volt power brick? It will fry at 43/19=2.25 amps. Actually, it will fry before that, even with a really good heat sink.
Consider buying a quality stepper driver like a Gecko (175watts $89ea x3) or a Linistepper (75watts $25ea x3).
-
While I agree with the cautions that James emphasizes, I'm not sure that I agree with the power dissipation calculation above - discussion on the other thread at [URL="http://www.cnczone.com/forums/stepper_motors_drives/139306-how_can_any_tb6560_driver.html"]http://www.cnczone.com/forums/stepper_motors_drives/139306-how_can_any_tb6560_driver.html[/URL]
-
Has any of you put a Short_Circuit_Protection circuit on your boards ?
You put a 0.01- 0.05 ohm resistor on the incoming power on the high side and use a pnp transistor that trips at -0.6 volts across the resistor, shutting down the enable on the chip. (Set the current trip to about 2X the max normal current) Use a R_S latch to keep it tripped.
This will also protect against any high current, assuming the disable will shut down the chip. Or you could use a Pchannel Mosfet as the switch to shut down power to the chip.
What really tends to blow power stages is if the PWM clock stops, which causes the upper mosfets to burn up quickly if they are partially on. This happens because the charge pump discharges. So you could monitor the CP oscillator and disable the chip unless its stable.
Larry K
-
[quote=James Newton;1097182]
it handles LESS than 43 watts (43 watts with an infinitely good heat-sink, which doesn't exist) which is only a bit over 1 amp at the max 40 volts. Even with a 24 volt supply, it will fry above 43/24=1.8 amps. Only using 19 volt power brick? It will fry at 43/19=2.25 amps. Actually, it will fry before that, even with a really good heat sink.
.[/quote]
I'm no expert, only a electronic tinkerer, but I don't believe your analogy is how it works. The spec you are referring to is Pd, which is the heat dissipation capability of the TB6560, but not all of the power is dissipated by the TB6560. If it was, then no power would be getting to the motor and no heating of the motor would happen. Say you are running 36V at 3.5A which would be 126 watts. Most of this 126 watts is going to go to the motor, not get dissipated as heat by the TB6560. How do we know how much the TB6560 really is dissipating? I'm not that smart, but I don't think it is blocking all 126 watts in my example, or anywhere near that amount for that matter.
I don't know what is exactly inside of a TB6560 chip, but I imagine the heart of it is some transistors. I don't have a lot of experience with transistors, only having used some in some circuit kits I've built. So I called my brother in law who repairs telecommunications equipment and works with them every day. His dumbed down description for me is that transistor has 3 basic regions, cut off (off), a linear region, and saturation (fully on). In the linear region, the transistor can make a lot of heat due to power dissipation, but if either fully on (saturated) or fully off (cut off), not much heat is created because the transistor is only acting like a switch. The linear region is where all of the heat is produced. He said that many times, transistors and are connected so that they operate in the cut-off or saturation regions only and only pass through the linear region very quickly. To further dumb it down for me, he used the rotary variable speed switch for a ceiling fan analogy. When the switch is off, there isn't any current flow so there can't be any heat generated. At low, or mid speed settings, the rotary switch has to regulate the power going to the fan so the fan runs but the switch has to dissipate the power and the switch gets warm. At high speed, the rotary switch is just passing all of the power that the fan can take, like an on-off switch. In this case, the switch also does not heat up.
With our stepper motors, each phase is being switched on and off very quickly but they are on or off, not partially on. This is how I understand it anyway.
What I believe you are doing is misinterpreting one spec in the datasheet and ignoring all of the others.
-
Hi 109jb,
You may be right, as the datasheet is a bit confusing to me on this point and I am also not really an /analog/ electronics expert. And even electronics experts get confused on power and heat issues. I'm also used to the linear design of the Linistepper driver which does dissipate any power not sent to the stepper rather than switch (which has it's own advantages) so this chopper type design may work differently.
But I don't see anywhere on the datasheet where it says that it can manage max volts at max amps. And I don't see anywhere else on the datasheet where it gives some idea of how much actual power (in watts = volts * amps) the chip is actually rated to supply.
So let me put it this way: Has anyone ever run this thing at 40 volts and 3 amps = 120 watts for any length of time?
-
[quote=James Newton;1097655]
...So let me put it this way: Has anyone ever run this thing at 40 volts and 3 amps = 120 watts for any length of time?[/quote]
I agree that running at 40V and 3A may overreach the capability of the design. On my machine I plan to use 3A at 24V (72 watts) so I will let everyone know how that goes.
On the power on sequence. I totally agree about that and the all-in-one boards with the BOB integrated. No real way to do it correctly with those boards from what I have read. For that reason, and because I don't like the idea of having to replace everything in the event of a failure of only a portion of the board (ie: just one axis fails, or just the BOB fails), I am opting for the single axis boards and a separate BOB. I will also have a separate power supply for the logic portion of the circuit. At first, I will just have to be careful about powering the circuit, but later I plan to have the motor power on a delay circuit so that it comes on a few seconds after the logic power.
My BOB and TB6560 single axis boards will be here later this month, so I will let everyone know of either success or failure when I can. Got to get the ballscrews in the machine first though.
-
[quote=datasheet]
High output withstand voltage due to the use of BiCD process:
Ron (upper and lower sum) =
TB6560AHQ: 0.6 Ω (typ.)
TB6560AFG: 0.7 Ω (typ.)
[/quote]
Dissipation @3.0 amps/phase would be
- (2 uppers on @ 0.7ohm) 12.6 watts
- (2 lowers on @ 0.6ohm) 10.8 watts
= 23 watts. (plus control circuit current)
Which is not much power, if properly heatsinked
The thing thats probably blowing your chips is current spikes (logic problems), or poor pcb layout creating bad current control or your pwm clock unstable, most likley.
-
[quote=Larken;1097750]Dissipation @3.0 amps/phase would be
- (2 uppers on @ 0.7ohm) 12.6 watts
- (2 lowers on @ 0.6ohm) 10.8 watts
= 23 watts. (plus control circuit current)
Which is not much power, if properly heatsinked
The thing thats probably blowing your chips is current spikes (logic problems), or poor pcb layout creating bad current control or your pwm clock unstable, most likley.[/quote]
So what you are saying is that the power (heat) that the chip has to dissipate is only 23 watts at 3 amps and is due to the internal resistance. This is not all of the power as most of the power just passes through and goes to the motor. Correct??
-
[quote=Larken;1097750]Originally Posted by datasheet
High output withstand voltage due to the use of BiCD process:
Ron (upper and lower sum) =
TB6560AHQ: 0.6 Ω (typ.)
TB6560AFG: 0.7 Ω (typ.)
Dissipation @3.0 amps/phase would be
- (2 uppers on @ 0.7ohm) 12.6 watts
- (2 lowers on @ 0.6ohm) 10.8 watts
= 23 watts. (plus control circuit current)
Which is not much power, if properly heatsinked
The thing thats probably blowing your chips is current spikes (logic problems), or poor pcb layout creating bad current control or your pwm clock unstable, most likley.[/quote]
I don't understand how you are arriving at 23 watts.
The TB6560AHQ is just a different chip package than the TB6560AFG. The AHQ is the one that is used on the "blue" boards. You can not add those dissipations. Ron for a single chip package needs to be multiplied by the motor current squared, times two for two windings being driven.
Also note that Ron shown in the datasheet at [URL="http://www.toshiba-components.com/motorcontrol/pdfs/TB6560AHQ_AFG_E_2003_20080407.pdf"]http://www.toshiba-components.com/motorcontrol/pdfs/TB6560AHQ_AFG_E_2003_20080407.pdf[/URL] for the AHQ is given on page 8 as 0.3 ohms typical (0.4 ohms max), and so the 0.6 ohm figure above already has multiplied that value by two (which is the meaning of "upper" and "lower", if I am understanding it correctly). However it could be that the "upper and lower" refers to the bridge circuit shown on page 24 of the datasheet, wherein current flows through two devices.
-
[quote=109jb;1097769]So what you are saying is that the power (heat) that the chip has to dissipate is only 23 watts at 3 amps and is due to the internal resistance. This is not all of the power as most of the power just passes through and goes to the motor. Correct??[/quote]
Yes, power to the motor= voltage x amps, so at 30volts @3 amps/phase the output power to the motor is 180 watts, but the chip is just disipating 23 watts of heat.
The upper and lower resistance refers to the On resistance of the Upper and lower mosfets in the bridge.
[quote=Doornob]The TB6560AHQ is just a different chip package than the TB6560AFG. [/quote] Ok, i thought it was saying upper and lower. '
So the power would be 25.2 watts for the AFG and 21.6watts for the AHQ.
-
[quote=Larken;1097809]Yes, power to the motor= voltage x amps, so at 30volts @3 amps/phase the output power to the motor is 180 watts, but the chip is just disipating 23 watts of heat.
The upper and lower resistance refers to the On resistance of the Upper and lower mosfets in the bridge.
Ok, i thought it was saying upper and lower. '
So the power would be 25.2 watts for the AFG and 21.6watts for the AHQ.[/quote]
Thanks. Makes me feel a bit better about my purchase.