TB6600 drive from EBAY

[A_Camera]

Hello... just before your post, I replied trying to give some info on the "New" N6 rev of these ebay drivers. I am in process of disassembly of one of these drives.

I'll wait for Admin to reply as maybe they don't want the info ???

In your post above you say the the LS14 is no longer part of the card. That is an improvement, not a degradation. If there is an inverter it should be a HC14, not LS14.

Attachment 269504
Attachment 269502

Even the cheap Chinese blue cards has HC14 inverters. Anyway, the TB6600 inputs don't need that, especially if you have high speed opto couplers, not those slow PC817, which is used on the blue card.

One thing which is definitely wrong with this card is the use of those five diodes. They should be removed and one of them replaced by a fast fuse.

Attachment 269506

Anyway, why wait for the Admin to reply? Of course they don't mind you publishing your findings, even if they are not personally interested. There are other members who would appreciate to see your findings.

[lonewolf55]

In your post above you say the the LS14 is no longer part of the card. That is an improvement, not a degradation. If there is an inverter it should be a HC14, not LS14.

Even the cheap Chinese blue cards has HC14 inverters. Anyway, the TB6600 inputs don't need that, especially if you have high speed opto couplers, not those slow PC817, which is used on the blue card.

One thing which is definitely wrong with this card is the use of those five diodes. They should be removed and one of them replaced by a fast fuse.

Anyway, why wait for the Admin to reply? Of course they don't mind you publishing your findings, even if they are not personally interested. There are other members who would appreciate to see your findings.

well First off Thanx Admin for finding my missing post and posting it

the reason to wait for Admin was out of Respect to the Rules of these forums.

I mentioned 74LS14 because that is what is listed on the schematic for the haoyuelectronics Module, that schematic shows no such Diodes, also the Schematic has no Revision Level.

I do not know what components are on the N2 Rev board because I do not have one ... I only have the N6 board.

the last image that you posted looks like the N6 but the area where the PCB numbers are on the upper left corner are not listed on your Image.

The Diode on the left, which I assume you wish to replace with a fuse is to prevent damage if the supply voltage polarity is reversed, the other 4 diodes are "FreeWheeling Diodes" put there as Spike Eaters, I can't say that is a bad Idea.

this was just an update to say Thanx for posting my original Post, I will post again when I have finished tracing out this N6 PCB

EDIT:
the Slow PC817 OC is just for the Enable signal, there is no reason it needs to be a Higher Speed OC, the other 2 OC's for Direction and Pulse are in Fact High speed units.

[lonewolf55]

these are the correct settings for the Step Settings on the N2 boards



NOTE: Switch in Down or ON position = LOW

[iamadude]

Hi everyone,

I have recently purchased the HY-DIV268n 5a driver (didn't realize it had the design faults mentioned here) and I was wondering if you guys can help me out in understanding some of the discussions in this thread (I don't have much experience working with electronics). I'm going to try to summarize some of the flaws and workarounds; if I'm misunderstanding or missing something please let me know.

So if I understood the discussion correctly, the HY-DIV268n driver only supplies 30% of rated torque because of the idle current reduction. This flaw is due to the low time constant and that the driver basically goes straight to the idle state. I see that there are a couple of suggestions to workaround this and these methods basically make pin 3 HIGH at all time (therefore at full torque at all time).

The workarounds I've seen are:
1. Solder pin 3 and pin 4
2. Solder pin 3 and Vreg
3. Remove pin 3, add a 10k resistor and supply 5 V

It seems the basic idea is to supply 5V to make pin 3 HIGH, but make sure there's a resistor in there as well.

The first method I saw here, the second and third methods I saw on other threads. For the first method, I've read the TB6600 datasheet but I’m still a bit uncertain about what pin 4 does exactly. I think pin 4 is always output HIGH (5V) unless the temperature is high, at which point, it outputs LOW for the TSD function. Since pin 4 is always HIGH, and it already has a resistor, you could short it to pin 3 to produce full torque. I think second method work as well since it's suppose there supposed to be a pull up resistor with the Vreg. I suppose it is the similar reason for the third method, add a resistor and keep supplying 5V to pin 3 for full torque.

Question - Is there a particular method that will work best? I'm thinking of just soldering 3 and 4 together.

I understand that even though these workarounds will work, they do heat up the driver and stepper motor considerably and should be carefully monitored for people's own applications. I'll probably add in some thermal paste to the heat sink since apparently they're not in there.

The other flaw I saw mentioned here is that the microstepping switches are mislabelled. I read that it's just SW column 4 and 5 should be reversed (so if the driver cover tells you to turn off column 4, you actually turn off 5 instead; similarly, if cover tells you to turn off 5 you actually turn off 4). Can anyone confirm this?

Thanks so much!

[H500]

The mod might depend on which version you have. When in doubt, the safest bet is to disconnect pin 3 from the board and then connect it to 5v. If you want to use pin 4, measure it to make sure it is connected to 5v. The latch pin is an input. It controls whether the drive will automatically reset if a fault condition causes it to shut down.

[lucas]

I mentioned 74LS14 because that is what is listed on the schematic for the haoyuelectronics Module, that schematic shows no such Diodes, also the Schematic has no Revision Level.

I do not know what components are on the N2 Rev board because I do not have one ... I only have the N6 board.

The schematic for the N2 board is attached at post 5 of this thread and it doesn't have a 74xx14, only transistors.
The N6 looks similar to the N2, you could use my schematic as a guide.

[lonewolf55]

The schematic for the N2 board is attached at post 5 of this thread and it doesn't have a 74xx14, only transistors.
The N6 looks similar to the N2, you could use my schematic as a guide.

Thank you so much Lucas, I have read this entire Thread word for word at least 2 dozen times LOL
I see the PDF now just looks like a Black Block in my Browser under your Thumbnails of the Oscilloscope WaveForms

from what I have Traced out so far this N6 that I have looks very much the same as the N2 board that you Traced out... there are however different component values for the resistors.

I will post my Schematic with the values that I discover as soon as I finish with it.

Also just a Note here, the Data Sheet for the TB6600HG is now on the Toshiba website.
easy enough to find at many places, but I do like from the Manufacture when I can get it

http://www.toshiba.com/taec/componen...E_EN_31288.pdf

[iamadude]

Thanks for the suggestion and explanation of the latch pin! I'll make sure to check pin 4's voltage before soldering to pin 3.

[A_Camera]

I see the PDF now just looks like a Black Block in my Browser under your Thumbnails of the Oscilloscope WaveForms

Click on the black thumbnail box and the PDF file will open.

[A_Camera]

The schematic for the N2 board is attached at post 5 of this thread and it doesn't have a 74xx14, only transistors.
The N6 looks similar to the N2, you could use my schematic as a guide.

Hi Lucas,

Thank you for the schematic. On that drawing you indicate that there are 330k pull down resistors on the inputs M1, M2, M3,CLK, CW/CCW and ENABLE. I don't think that is a good idea since the chip has 100k pull down resistors built in, so the two in parallel results in 76k, which is significantly less than the 100k. I'd remove all the 330k since they do no good and in my opinion pointless, just like the diodes on the motor outputs. I can understand some use for the diode on VMA and VMB, but a fast fuse would be more important.

[lonewolf55]

Looking at the Data Sheet ...

[3] If your design includes an inductive load such as a motor coil, incorporate a protection circuit into the design to
prevent device malfunction or breakdown caused by the current resulting from the inrush current at power ON
or the negative current resulting from the back electromotive force at power OFF. IC breakdown may cause
injury, smoke or ignition.

since many claim these 4 diodes on the output lines do Nothing, obviously they cause no Harm either. IMHO I suggest they be left as is

as for the the other diode, it could be replaced with Fast Acting Fuse, but then again if the fuse was rated high enough to not be blown by inrush from Power on, it may be ineffective ... a good solution might be a slow turn on supply, but here again we are increasing Costs.
Still I think a Fused input for this N6 Driver Module is a Very good Idea.

[lucas]

Thank you for the schematic. On that drawing you indicate that there are 330k pull down resistors on the inputs M1, M2, M3,CLK, CW/CCW and ENABLE. I don't think that is a good idea since the chip has 100k pull down resistors built in, so the two in parallel results in 76k, which is significantly less than the 100k. I'd remove all the 330k since they do no good and in my opinion pointless,

The schematic is there since the beginning and proves the bad design quality.
Those 330K are useless indeed, shows how well the designer did read the datasheet...
Removing these is difficult without proper equipment and not damaging the PCB.

I can understand some use for the diode on VMA and VMB, but a fast fuse would be more important.

The diode in series with the supply MUST go, it prevents the back EMF from a decelerating motor to be dumped in the power supply.

Looking at the Data Sheet ...

[3] If your design includes an inductive load such as a motor coil, incorporate a protection circuit into the design to
prevent device malfunction or breakdown caused by the current resulting from the inrush current at power ON
or the negative current resulting from the back electromotive force at power OFF. IC breakdown may cause
injury, smoke or ignition.
....

as for the the other diode, it could be replaced with Fast Acting Fuse, but then again if the fuse was rated high enough to not be blown by inrush from Power on, it may be ineffective ... a good solution might be a slow turn on supply, but here again we are increasing Costs.
Still I think a Fused input for this N6 Driver Module is a Very good Idea.

That diode must go, see above and the datasheet quote on negative current at power OFF,
Inrush current will only occur when you switch the DC supply, normally the supply is switched at the mains site and the DC output will rise slow enough to limit the inrush current.
A fuse is always needed, must it be on the drive? I prefer it on a power distribution assy, some like it chassis mounted.

since many claim these 4 diodes on the output lines do Nothing, obviously they cause no Harm either. IMHO I suggest they be left as is

Those 4 diodes should reduce the chip power dissipation in slow decay mode and are not there for spikes.
I did measure the effect on the THB6064 chip and the effect was marginal, The TB6600 is different in decay settings, it needs testing to check if they do something.

I never tried to make a "how to modify this drive in a good one", because there are so many things wrong with it:
Mechanical design, component quality, see this thread for more.
The small SMD makes it difficult or impossible for most people.

[A_Camera]

Looking at the Data Sheet ...

[3] If your design includes an inductive load such as a motor coil, incorporate a protection circuit into the design to
prevent device malfunction or breakdown caused by the current resulting from the inrush current at power ON
or the negative current resulting from the back electromotive force at power OFF. IC breakdown may cause
injury, smoke or ignition.

Yes, that's generally true, but the TB6600, and also the older TB6560 contains those diodes already, so adding more diodes won't increase the protection, quite the opposite, if those added diodes are slow then your protection level is lowered. It is at least my understanding of the data sheet that those diodes are built in and not needed to be added.

since many claim these 4 diodes on the output lines do Nothing, obviously they cause no Harm either. IMHO I suggest they be left as is

To me those diode makes no sense at all.

as for the the other diode, it could be replaced with Fast Acting Fuse, but then again if the fuse was rated high enough to not be blown by inrush from Power on, it may be ineffective ... a good solution might be a slow turn on supply, but here again we are increasing Costs.
Still I think a Fused input for this N6 Driver Module is a Very good Idea.

The diode on the power may indeed may provide a level of protection against wrong polarity, but how often do you connect and disconnect the card? My guess is that once you are up and running you never disconnect it again. Even so, careful design of your connection should prevent wrong polarity so leaving that diode makes no sense. Regardless of which, a fast fuse is DEFINITELY advisable to add and that is also clearly stated in the data sheets.

No external diodes are mentioned in the data sheets, quite the opposite, it is clearly stated that these are built into the chip.

[A_Camera]

The schematic is there since the beginning and proves the bad design quality.
Those 330K are useless indeed, shows how well the designer did read the datasheet...
Removing these is difficult without proper equipment and not damaging the PCB.

If you don't want to save for future use then it is fairly easy to cut them in the middle if you can't de-solder. I don't have that card, so I don't know how they are placed but I'd definitely do something about those resistors. At least on my card, the infamous "Blue card", they used 6.8M pull downs, so they don't have such high influence. Even so I removed also those.

The diode in series with the supply MUST go, it prevents the back EMF from a decelerating motor to be dumped in the power supply.

That diode must go, see above and the datasheet quote on negative current at power OFF,
Inrush current will only occur when you switch the DC supply, normally the supply is switched at the mains site and the DC output will rise slow enough to limit the inrush current.

Ooops... sorry, you are of course right. Not only they are pointless but they provide a false sense of protection.

A fuse is always needed, must it be on the drive? I prefer it on a power distribution assy, some like it chassis mounted.

No, the fuse must not be on the drive but if it was there people would be aware of it. Now that there is a diode, people think it provides similar protection and ignore the need of a fuse. I have also added external fuses, it is easier to change in case they need to be changed.

Those 4 diodes should reduce the chip power dissipation in slow decay mode and are not there for spikes.
I did measure the effect on the THB6064 chip and the effect was marginal, The TB6600 is different in decay settings, it needs testing to check if they do something.

OK, but I am not convinced they have any use at all, though I am convinced that if they are there they should be fast diodes, not cheap slow ones.

I never tried to make a "how to modify this drive in a good one", because there are so many things wrong with it:
Mechanical design, component quality, see this thread for more.
The small SMD makes it difficult or impossible for most people.

I know that it is not easy to work with SMD unless you have the right tools, but even without right tools, if you are handy you could modify and correct the major errors. Maybe if I buy this driver I'll document the changes and start a "How to fix" thread. Generally, I don't think it should take more than an hour to fix the card for anyone with some soldering experience, so once it is documented it is a fairly easy job for most DIY CNC builders. The price of these cards is low, so adding an hour to each card makes it still worth the effort. Of course, documenting the needed changes, doing all the measurements and testing all the corrections takes time, but that's a different matter.

[fluvannabear]

A_Camera: I have been reading all the posts I can find on the TB6600 Driver. Wow, so many things wrong from the factory. I am trying to put together a 4 axis system for my G0484 Mill/Drill using the 4 axis version of that driver and NEMA23, 425oz-in Motors with a 350W-36V power supply. StepperOnline has this package at what seems a reasonable price. Therefore I am really looking forward to your "How to Fix your TB6600 Driver" write-up. Hope you have a chance to begin the effort. Take care. Burt

[109jb]

A write up on fixing the single axis driver probably won't help much for the 4-axis board. It could have a totally different set of problems

[fluvannabear]

109jb: Well that's no fun.. Don't know enough electronics to fix it myself. Grew up during vacuum tube days. They made sense. Well, may just have to bite the bullet and try it anyway. Hope get more replies that may address the multi-axis drivers using the TB6600 design. Thanks and take care. Burt

[H500]

I am trying to put together a 4 axis system for my G0484 Mill/Drill using the 4 axis version of that driver and NEMA23, 425oz-in Motors with a 350W-36V power supply. StepperOnline has this package at what seems a reasonable price.

Their prices looks good, but I'm guessing that courier shipping and brokerage from China will be very expensive.

Although multi axis boards are slightly easier to set up, they are much harder to troubleshoot and fix if something goes wrong. You can't simply swap the damaged channel with a spare. Also, kit components are often not properly matched for best performance.

[A_Camera]

A_Camera: I have been reading all the posts I can find on the TB6600 Driver. Wow, so many things wrong from the factory. I am trying to put together a 4 axis system for my G0484 Mill/Drill using the 4 axis version of that driver and NEMA23, 425oz-in Motors with a 350W-36V power supply. StepperOnline has this package at what seems a reasonable price. Therefore I am really looking forward to your "How to Fix your TB6600 Driver" write-up. Hope you have a chance to begin the effort. Take care. Burt

Hi,

I am sorry, but I will not buy a multi axis driver again. I have two such cards using the TB6560 and fixed that once. They are identical (that's what I thought), so the second one was just to copy the changes from the first. In fact, I fixed the second one yesterday and seen that the second is not 100% identical so it is a bit different version from the first, but the major fixes were the same. I also learned a lesson, which is never to buy a multi axis card again. It is much easier to work with single axis cards than multi axis. As far as I am concerned, the only advantage of multi axix cards is the space (need less space and wiring) and the price (cheaper), but the price difference isn't worth the extra work when fixing is needed, and also if one channel is damaged you have to throw everything away unless you are ready to spend a whole lot of time removing and replacing the broken parts.

Another thing I learned...

I would NOT buy another stepper controller card from China if I didn't have the electronic knowledge, experience and all the instruments I have. I would also NEVER advice anyone to by one of those unless they have at least basic electronic knowledge, soldering experience and soldering tools and at least some measuring instruments. Those cards use surface mounted parts and those are not always easy to remove and replace with the right values.

When I bought my card I believed it was just "plug and play", but... although I knew that if get into trouble I could fix it and also find it interesting to fix electronics, so I was not worried. Never the less, I expected "plug and play" so I was not very happy discovering all the issues. Even worse is the fact that I found this site too late, when I already fixed my card. If I'd have found this place earlier I wouldn't have bought that card at all. Now it is a sort of obsession to start using it. I burned the first one last week and now I fixed the second one and also interfaced to a USB controller. I will see tonight if it spins the motors. If not then I'll buy three TB6600, or TB6064 drivers and sort of start over. It is very irritating to end up with these activities just when I was ready with the mechanical work and my CNC is to 99% built, the only still missing parts are limit switches on X and Y and also the protection cover I plan to have around it.

[lonewolf55]

A write up on fixing the single axis driver probably won't help much for the 4-axis board. It could have a totally different set of problems

I will be uploading the schematic for the HY-DIV268N-5A Rev N6, I'm almost finished with it, I can tell you right now that it is pretty much the same as what Lucas has discovered.
I may or may not mention all my findings, but suffice it to say, the best approach would be to Design a new PCB.
more on that when I post the schematic

Thanx much for all the help here.