bipolar stepper drive- uc3770 or discrete method...

[jaydag71]

I am trying to drive a bunch of those tiny bipolar steppers salvaged from 3.14floppy drives- they are pretty small 5V 2phase. I have a bunch of uc3770b stepper drive ic's but haven't been able to figure out how to hook everything togeter yet- confused about what to feed for a voltage ref, how to implement the phase, and some other parts are fuzzy too- I have enough parts around that I could assemble a bunch of hbridge transistor circuits which I'm sure would be a good fit but still confused about how to use phase with the discrete hbridges amoung other things....

The goal is to drive these steppers back and forth only a few degrees, say 90 forward then 90 back, in various looping patterns. Any help or tips to online resources would be great- I've been looking for days but not any luck for the IC and nothing simple for the discrete circuits.

 

[gramo]

The L297 and L298 are a great combo for bipolar steppers!

**broken link removed**

**broken link removed**

I think the L298N has a built-in H-Bridge (all the diodes shown above)

 

[jaydag71]

Thanks for reply- yes aware of the combo- no more cash and canot get them thru sample prog anymore! would be ideal to use due to amount of coverage online and snippets to be hacked but oh well. Not committed to the drivers I have now but would like to figure them out for testing the steppers at least- see if they are going to be suitable for the design.

uc3770 x2 is said to be similar to the Allegro 2916- I have been trying to use portions of their example design and just readin through the block diagram of the 3770 datasheet but it really doesn't provide much help- surely someone who knows better how to interpret the diagram could shed some insight into the use of the device?

I want to use a pic16f to drive the test circuit for now.If I have to build some hbridges I have some designs but need the portions which would protect against short input combos using transistor rather than logic IC- Basically if I was going to go with a discrete design I would need it to be optimized for the required voltage, cheap to implement and small in board realestate cause I would need so many!

anyways gotta go to bed, not even making much sense anymore will write more tomorow.

 

[jaydag71]

still in need bipolar stepper helpz!!! help.?

I have slept on the matter and still am in need this morning!

As far as the driver IC is concerned, again I think I would be good if someone who is better able to interpret the datasheets's block functional diagram could shed some insight into the particulars of the pins which are giving me trouble?? Specifically I am confused about: what to tie the phase pin(s) to (I assume it owuld be the uC or controlling circuit but am fuzzy about what I should issue- hi for what/low for what? The emmiter pin(s) I would imagine are tied to ground, no? (because the VM would be tied to the motors voltage source?) What about the timing pin(s), do these get their signals from the controlling device/circuit? or is there supposed to be a dedicated source for these signals designed? Does the current pin(s) get tied directly to vcc or are there circuits which are intended to alter the signal prior to driver? Finally the voltage ref pin(s) are supposed to be tied to one of the voltage sources- but which one? motor volts or logic volts?

Again the datasheet is very limited in coverage- it provides only the block diagram and maximum ratings breakdown- both of which are very helpfull but obviously not enough for my skill-level...........

any help with interpreting the device would be 1st on my wish list for today...secondly if a discrete solution might be a better fit and anyone can point me to a good design source which addresses short-combo protection using transistors rather than logic IC's that would be great also!

The key to the drive- either method -is to keep board size super little and lite! If I could make use of SMT transistors with a crafty layout which would amount to less area than dedicated IC's then I would rather go that route!

The current drivers I am using are only for testing the steppers- due to their inherit size restrictions (2 per motor) I wouldn't use them in final design ever. One other set of IC's I have handy are UN2803 darlington transistor arrays. from what I understand of them they are better suited for unipolar steppers- but in looking at their makup I wonder if they coult just as well be used for a bipolar stepper, even two per IC? Each device has 8 darlington pairs, stepper has four per- from what I understand the issue would be in controlling the transistors 'firing' ??Is this wrong? I cannot find any ref designs implementing bipolar steppers using the devices but it seems doable?

Bottom line I badly need some sort of tips to get a working design going for testing stage!! Still dunno if these steppers are going to be suitable for project!!

 

[philba]

it would help to provide a link to the datasheet.

I wouldn't use the 298/297 combo for those motors - that is vast vast overkill.

The darlingtons aren't a good fit for bipolar motors because you need both highside and lowside drivers. However, you can easily convert a bipolar motor to unipolar - look at the wiring diagrams for each and it should be obvious. google for "jones stepper" for a great tutorial. Just remember that you get a pretty significant voltage drop with the 2803 so might want to increase the supply voltage a bit.

 

[gramo]

However, you can easily convert a bipolar motor to unipolar - look at the wiring diagrams for each and it should be obvious.

Downside to this is loss of torque, a lot of it.

 

[philba]

There is some loss but there are several techniques to over come that. Frankly, those motors (floppy drive) aren't going to be very powerful at all. I was responding to his desire to use what he has.

There are several inexpensive solutions to driving small bipolar steppers. Jones has lots of examples. If the current draw is low enough, you can use 74XX logic (I'd use 74HC) or an L293D (<$3) or the [SIZE=-1]TI SN754410 (<$2) You will need to drive those with some logic for stepping. A small micro works great. Again, Jones on Stepping is a great source for circuits.[/SIZE]

 

[jaydag71]

Thanks to all who have replied! I have made some headway since last post...

firstly thanks for the unipolar conversion suggestion- I won't do it because I believe that I will need every speck of torque I am able to obtain from the stepper- must admit that I had not thought of that approach though!

I have been considering using logic IC's as the drive circuit but generally they are more complex than the integrated solutions for me to use- being that I still am struggling with circuit design basics I think that using a dedicated IC would be the best way to go in order to actually produce something rather than just learn something!

Also partially for that very reason I would be using the L293D driver solution- it will take up alot less board area when considering driving 16-24 motors in final design- easily swappable when using sockets in case devices do burn out (and they will...) Device is recommended by stepper manufacturer based on similar specs given for similar stepper., and much more online support due to much more general use.

Somehow I managed to find 1 single L293N device this afternoon while doing a bit of bench cleaning- I am going to forsake the uc3770 device(s) and try using the 293N- will likely be running quickly....

Anyways, about the uc3770b, (for those who will inevitably come across this problem at some point in your travels....) after spending much of yesterday doing much more googeling I managed to find a couple of deeply buried threads from other discussion boards- one of which mentioned the original IC manufacturer making a merger with another- long story short managed to find out that New Japan Radio manufactures and supports the IC. Link:
**broken link removed**

The site also maintains a bunch of application notes related to the device and similar. There is a few example circuit/typical circuit schematics etc. So I managed to get the devices mostly wired up to the stepper- still a bit confused about implementing the sense circuit for my motors specs but the datasheet does provide basically everything needed to figure these values out I believe... so I shoudl be able to get the devices running if I wanted to, but again I'm just going to forget about them for now and go with the 293N due to its ease of use..

Thanks again to all who have helped out! jd

 

[gramo]

Great choice on the 293, that one slipped my mind

Its great for low power motors, as it can handle up to 600mA per channel (1.2A peak) But it sounds great for your application.

**broken link removed**

As you can see, it has the built in diodes for Back EMF protection already, so that really helps to keep external parts to a minimum!

 

[philba]

the L293 doesn't have the diodes but the L293D does.

On the 3770, I think it wont really help you any more than the L293D because the 3770 is only a single H Bridge. You need 2 3770s for each stepper and you will still need to generate the step signals. By the way, the SN754410 is pin compatible with the L293D including diodes.

Mea culpa. I was completely confused earlier - you can't make a unipolar out of a bipolar. It's the other way around. sigh...

On a somewhat different topic, how much force do you need from those steppers? they are not very powerful at all. what is the coil resistance?

 

[Dr-Chambers]

Hows everybody doing?

First of all Jaydag71 takes for the website you put up.

I am doing something like Jaydag71 but I need the bipolar stepper motor to move in both directions(to be bidirectional).Therefore the conversion would not be suitable in this case as you already know.

I am planning to use the LMD18200T T0220 as the H-Bridge driver to drive the bipolar stepper motor(Is that a good idea?).The bipolar stepper motor will be used in the application of process tomography where it will drive an optical sensor array up and down a pipe for a certain number of times to look inside it.
I was hoping if someone could help me with the H-bridge drive circuit schematics. The software that I will be using is Matlab using Data Acquistion card.

Thanking you in advance.Thank you.