H-Bridge HIP4081A Question

I am trying to follow the following tutorial
http://itp.nyu.edu/physcomp/Labs/DCMotorControl

I have all of the parts but the H-Bridge, I can buy the same one(Texas Instruments L293NE or Texas Instruments SN754410) they have off Ebay but it will take weeks to get to me and I will have to buy 5 of them.
Another option is my local electronics store sells one H-Bridge but it’s not the same, it’s a HIP4081A High Frequency H-Bridge Driver. I read through the datasheet but it’s very different from the one they used and I am pretty new to electronics and diagrams and don't understand the pin out descriptions. I hoping someone can tell me if it’s similar enough to work for that tutorial and maybe give me a pointer or two on how to hook it up?

You can find links to the datasheets below.
HIP4081A
http://www.robotpower.com/downloads/HIP4081A-datasheet.pdf

SN754410

Or
L293

Thanks in Advance for any Help.
Chris

The HIP4081 is used to drive 4 large n-channel power MOSFETs in an H-bridge configuration. Such an H-bridge can power large motors (100's to 1000's of watts), but need careful design and testing. Its not really a beginner's project - there is an open-source motor controller project (OSMC.org I think) which has designs using the HIP4081.

The good things the HIP4081 brings to the party are: configurable shoot-through prevention, bias voltage generation for high-side n-channel MOSFETs, high current gate drivers for high and low side, very high frequency operation, upto 80V supply voltage.

You can of course use the HIP4081 for a low-power H-bridge, but its not the cheapest/simplest option. I've used a LMD18200T for that before, which has everything in one device ( I think its 36V 3A or similar )

I'm really surprised that you can't find an L293 or equivalent locally (which is where, exactly?); they seem common as dirt to me (then again, I am in the States - there are probably surplus dealers here who have 55 gallon drums full of 'em).

Have you considered making your own h-bridge from some TIP120 transistors or similar? It would teach you a lot about how h-bridges work, plus the parts would probably be easier to source. If you use bipolar transistors, you can build an h-bridge from all NPN, or a pair of NPN and a pair of PNP (which is more efficient than an all NPN). Going with mosfets is doable, but an all N-channel h-bridge is difficult to create without a high-side driver or charge pump (thus, going with a pair of n-channel and a pair of p-channel works better - if you can find a complementary pair of p-channel mosfets).

Something approaching a single bridge of a L293 would be to use a pair of 2n2222 (NPN) and a pair of 2n2907 (PNP - complement to the 2n2222); this will run small DC hobby motors fairly well. You would of course have to manage things to prevent shoot-thru, as well as implement other controls (and you wouldn't have any form of overheating or overcurrent protection), but such a small h-bridge can teach you a lot about such design (and if you blow parts while building and testing it, you haven't lost a bunch of money).

There are plenty of tutorials and examples of such small transistor h-bridges out there on the internet - so consider it as a possibility, perhaps.

:slight_smile:

I would suggest if you aren't able to procure these under defined H-Bridges then try out making your own, here is a very good tutorial >> http://www.bristolwatch.com/ele/h_bridge.htm using power Mosfet's.

also even if you want to use small hobby motors like the ones futaba makes and are also found with TAMIYA hobby kits(i have one of them)then in that case also they call for a lot of Amperage to be handled you have to use the D -cell method to calculate the H-Bridge value before buying or even making one.

You have to go like 1) Hold the multimeter at DC amp level and put the + lead and - lead of the multimeter with your motor leads and power leads.

  1. Hold the movement shaft of the Motor from front and do not let it move and power it

while the multimeter is there it will show you the STALL current for the motors you have and as per that you have to choose the building mosfets or ready made H-bridge IC's for your implication telling you the highest current soaking that can happen in the worst case scenario's.

I used 4 L293D's by STACK SOLDERING them one over one and thus multiplying there original power.

EDIT: Example of Stacking attached for your reference.

NI$HANT:
I would suggest if you aren't able to procure these under defined H-Bridges then try out making your own, here is a very good tutorial >> MOSFET H-Bridge Motor Control using power Mosfet's.

That's not a bad tutorial; the only problem is sourcing the p-channel mosfets (can be done fairly easily if bought new from places like mouser and digikey, but for some reason, buying p-channels from ebay or surplus isn't as easy, at least ones matched in performance to the n-channels; I think this has to do with costs, as n-channel mosfets tend to be cheaper, plus for h-bridges at least, the easy availability of drivers/charge-pump ICs for all n-channel designs). But - if you can get the parts, they tend to be easier to build.

NI$HANT:
also even if you want to use small hobby motors like the ones futaba makes and are also found with TAMIYA hobby kits(i have one of them)then in that case also they call for a lot of Amperage to be handled you have to use the D -cell method to calculate the H-Bridge value before buying or even making one.

Or use a bench/lab power supply (for really large motors, you have to get creative - I have a car starter/battery charger that I use for anything needing 10 amps or more).

NI$HANT:
You have to go like 1) Hold the multimeter at DC amp level and put the + lead and - lead of the multimeter with your motor leads and power leads.

  1. Hold the movement shaft of the Motor from front and do not let it move and power it

If you're going to describe this process, be clear about it (otherwise damage to the meter, motor, or user is a possibility):

  1. Put your multimeter in current measurement mode; most meters can handle up to 20 amps, but check your meter (put the leads into the proper jacks on the meter, if the meter has separate high-current jacks).

  2. The meter needs to be placed in series with the load being measured: So, positive from supply to load, other side of load to positive of meter, negative of meter to supply ground.

  3. Mount the motor so that the body can't spin; grip the shaft so it can't turn (note: on larger motors, DO NOT try to do this with your hands/fingers; fabricate a pony-brake test stand instead).

  4. Verify that when power is applied, nothing is going to start spinning and fly off and whack you in the head.

  5. While reading the meter - apply the power BRIEFLY, and note the current measurement. Do this a few times, but each time very briefly.

The reasoning for step 5 is the fact that you are, in effect, "abusing" the motor and the meter; if you do this for more than a second or so at a time (and on larger motors, potentially less), you can easily burn out windings, and in the case of the meter, blow a fuse or otherwise burn out the meter (it's current measurement design isn't likely designed for constant high-current measurement in this kind of situation). Generally, you should do this kind of test with an el-cheapo throwaway meter from some discount tool vendor (like Harbor Freight), and not with your "good meter".

NI$HANT:
while the multimeter is there it will show you the STALL current for the motors you have and as per that you have to choose the building mosfets or ready made H-bridge IC's for your implication telling you the highest current soaking that can happen in the worst case scenario's.

Another method would be to build a pony-brake test stand with an adjustable slip fitting, such that the shaft can still turn, but just barely; this will give you a "heavily loaded" stalled reading; just add 10-15 percent for fully stalled (and that will also give you some breathing room, too).

NI$HANT:
I used 4 L293D's by STACK SOLDERING them one over one and thus multiplying there original power.

I've read of people doing that with two, but only guesses at doing it with four. To be honest, I'm surprised even two work. There are a couple of reasons why:

  1. Lack of cooling capability - L293s radiate heat from the center set of pins, and the top of the package; stacking them means the ones on the bottom can't effective cool themselves properly...
  2. BJT outputs - the L293 (and L298) using BJT (bipolar junction transistors) on the output stage; such transistors typically can't be paralleled properly (just like diodes), and the load won't be shared evenly on high currents. What happens then is a cascade failure. It is also why both h-bridges aren't very efficient, as compared to mosfet bridges (I'm surprised they've never release drop-in compatible mosfet designs for the L293 and L298)...

...and yes, I know it probably works for you, and its worked for others - why or how you get away with it, I don't know; I've never found a good explanation for it.

If you're going to describe this process, be clear about it (otherwise damage to the meter, motor, or user is a possibility):

I didn't write using hobby motors i assumed that fact , as per the tutorial he is following most probably the motors would be Hobby grade(its very rare but not impossible to destroy multimeter's with these type of motor's) not to expect a RC Heli Motor :sweat_smile: ,well you have been specific im fine with that.

stacking them means the ones on the bottom can't effective cool themselves properly...

I had a remedy to it too!, inserting a Aluminium plate that i custom cut from some scrap at my workshop and made sure it fitted in tight so the transference of heat to the plate happens and then giving the shape of a wing to the excess of plate getting out of the bottom IC's can be worked out.

NI$HANT:
I didn't write using hobby motors i assumed that fact , as per the tutorial he is following most probably the motors would be Hobby grade(its very rare but not impossible to destroy multimeter's with these type of motor's) not to expect a RC Heli Motor :sweat_smile: ,well you have been specific im fine with that.

Well - I suppose you could assume that; but others may read the process, and think they could apply it to a larger motor; and indeed, even my steps will only allow you to test only up to a certain size motor. In fact, I'm not even sure how you would test an unknown but very high-power DC motor (say perhaps a DC gear motor for a motorized wheelchair for which you don't have specs). You most likely couldn't pony-brake such a motor - not safely, anyhow. I wonder if anybody has suggestions?

NI$HANT:
I had a remedy to it too!, inserting a Aluminium plate that i custom cut from some scrap at my workshop and made sure it fitted in tight so the transference of heat to the plate happens and then giving the shape of a wing to the excess of plate getting out of the bottom IC's can be worked out.

Interesting - I know that there exist clip-on heatsinks for the L293 in its "standard" configuration. I still wish I had an explanation why or how it can work "stacked", since it has those BJT outputs (I've read numerous warnings not to try and parallel multiple BJTs, as they'll fail - and this would seem to be an instance of this)...

(I've read numerous warnings not to try and parallel multiple BJTs, as they'll fail - and this would seem to be an instance of this)...

Ya can be but still they are on but this doesn't even explain that the technique or use is fine as per the concepts.

Wow First thanks for the greats replies guys! I am indeed just playing around and trying to learn a bit as I go.
I have a simple FK-130RH motor
http://www.mabuchi-motor.co.jp/cgi-bin/catalog/e_catalog.cgi?CAT_ID=fk_130rhsh), very small.

I am up in Vancouver Canada and the only two electronics stores I know about are
Lees Electronics http://www.leeselectronic.com/product_info.php?products_id=3243
Main street Electronics http://www.mainelectronics.com/

I would like to attempt to build my own H-Bridge, as I think that would be a good project to learn from and I already have 2 TIP-122 NPNs, Again Overkill(100v vs 60v) but I went into Lees Electronics looking for TIP-120s and all he stocked was the 122s, he reasoning was it is a similar part and would cover more projects and allow him to only stock one. Anyway I got the 122s for $1 each. The other thing is there website says they stock the HIP4081A for $2 so I wasn’t worried about the price, until I looked it up on findchips.com and I see the average price more in the $7 range, so I think the website is wrong, I am planning on giving them a call tomorrow to see what they currently carry as I went in there once for a ATmega328 chip that was on their website and they did not carry it anymore as he said the ATTiny2313 out sold it so that was all he carried now.

Anyway I found this link, thank you NI$HANT for the link I did read through it and I do understand the diagrams but my ability to turn them into reality on the breadboard is still limited. The following one is a little better for a beginner like me.

I might have to just buy one of the L293s off of Ebay to complete that tutorial or learn to build my own first.

Again thanks for all of the help, any more suggestions are always appreciated.
Chris