Connecting l293d h bridges in parallel?
rwgast_logicdesign
Posts: 1,464
Ok I screwed up big time, Id like to use the chips I have but if not I really need the cheapest H-bridge solution..
I have two motors and like an idiot i tested them without a load, at no load they draw around .1 amps each, so I went ahead and ordered 3 293d chips thinking one would drive both motors...
Well I just tested my bot out with a meter and it looks like the motors are drawing just about .6 amps rolling on the ground with no obstructions, and soon as the wheels lock up the thing can draw up to 1.4 amps or so.... the chip says its max is 600 mA which is what I draw just crusing
Can I connect all three of my chips in parallel to get a working motor controller? At the moment I have no way to get more components, but when I do what would be a better options mosfets? I originally looked at the ST 298 chip but it doesnt have a very proto board friendly package. Im really hoping I can just hook up all 3 l293 chips to make one big motor driver at least for now
.
Also totally off topic but is wd40 bad for nylon gears? I noticed my pinion gears are cracking and I have no idea why I may have to replace my gears and would rather not have them crack again,
I have two motors and like an idiot i tested them without a load, at no load they draw around .1 amps each, so I went ahead and ordered 3 293d chips thinking one would drive both motors...
Well I just tested my bot out with a meter and it looks like the motors are drawing just about .6 amps rolling on the ground with no obstructions, and soon as the wheels lock up the thing can draw up to 1.4 amps or so.... the chip says its max is 600 mA which is what I draw just crusing
Can I connect all three of my chips in parallel to get a working motor controller? At the moment I have no way to get more components, but when I do what would be a better options mosfets? I originally looked at the ST 298 chip but it doesnt have a very proto board friendly package. Im really hoping I can just hook up all 3 l293 chips to make one big motor driver at least for now
.Also totally off topic but is wd40 bad for nylon gears? I noticed my pinion gears are cracking and I have no idea why I may have to replace my gears and would rather not have them crack again,
Comments
And H-bridge configuration just makes the whole idea of parallel a bit more complex. If you really need a powerful h-bridge, I've had good luck with Bob Blick's design and using darlingtons that are less than $1usd each.
http://www.bobblick.com/techref/projects/hbridge/hbridge.html
But there are even better MOSfet designs these day. I suspect the easiest way to drive the power MOSfets are with small transistors as MOSfet drivers are in SMT packages and both expensive and hard to get.
You might just be able to use the big brother of the L293, an L298. There are several limits involved in h-bridges. One that is important and least documented is the voltage drop of the h-bridge itself.
With TIP120/TIP125 Darlingtons that Bob Blick uses, the h-bridge voltage drop can be between 4 and 8 volts at saturation, so a 12 volt motor might require a 16-20 volt supply. Then there are 3 limits of stress to the h-bridge - maximum amps, maximum volts, and maximum watts.
With power MOSfets, those voltage drops tend to go way down. But since there are so many MOSfets, I can't give you an exact figure.
Having said all that, I don't really have a good reference for a MOSfet h-bridge that is completely proven for DIY.
two of these in parallel will pull 1.2 amps continous and 2.4 peak wich is almost double what im using the problem is im right over the max specs of 1 chip.
i also couldnt find a peak current time how long is it safe to peak, what if my bot is stuck on a wall for a few minutes? and what happens if u exceed peek rating for a few seconds, will the chip just smoke
I wish you luck with your current endeavor, but if that does not work out for you, the L6205 IC might be just the ticket for you. See the datasheet below. Parallax used this chip in a motor controller board a while back, but have since discontinued it.
STMicro offers free samples for three different forms. I believe they will allow you to get three samples at the most. Please refer to the following link:
Bruce
In any event, I have occasionally gotten free sample, but don't really care to abuse it due to the problem of international shipping costs.
http://forums.parallax.com/showthread.php?130758-Need-Help!-Can-SN754410-handle-40RPM-12V-22-oz-in-Torque-160mA-DC-Gearmotor&highlight=sn754410
I usually use one (or two) of these TI SN754410 chips for smaller motors, the next step up I use is the L298 chip and beyond that I really like the LMD18200 which works well for larger motors and higher voltages. Beyond that the next step is a custom H-bridge or an OSMC board.
Robert
Besides, if you really want to run a motor over about 40 watts, you are going to need something that is not in a chip. 1/10th of a horsepower is 74 or so watts.
I've pretty much finished with L293 and L298 devices - too small.
It all depends on the current and voltage requirements of the motors you are using. For smaller motors with little load the L293/SN754410 and L298 chips work well. For larger devices the LMD18200 is fine. Beyond that then H-Bridges made from MOSFETS may be the best option. That is what the OSMC motor controllers use. You can install up to 4 MOSFETS per leg.
Using the ready built H-bridge chips can be easier to use and wire up. The LMD18200 also has some built-in protection and when used properly is a pretty robust chip. If you decide to roll-you own H-bridge with individual MOSFETs in many cases you also need to provide another voltage source higher than your motor voltage in order to fully turn-on the high side MOSFETs. You could use something like a HIP4081 driver chip or use a small DC/DC converter to generate the higher voltage. If you aren't using a dedicated driver chip like the HIP4081 then you may also need to add some additional drivers so the microcontroller can switch the larger voltages to the MOSFET's. It is an interesting area to explore but unless motor control is an area that the person wants to get into it may be better to get an off the shelf H-bridge module when the need for larger motors comes up.
Robert
I have a question though my motor are 3-6v this chip says it needs a supply 8-52v, Does that mean its going to take my pwm signal and output it at 8v minimal??
Or you can order it directly from Seeed Studio
For more info look at this wiki page.
No, it means that the second power source is supposed to be in the range of 8V to 52V (max). You'll probably need a separate higher voltage battery pack for your motors. Or just use a larger pack overall with the direct battery power connected to the motor power.
Most of the H-bridge chips have logic level inputs but many of them are setup to use a separate voltage source for the motor outputs that is a higher voltage than the logic side. Some specify that the motor supply needs to be a couple volts higher. You'll need to look at the datasheets carefully to see if this is the case with the H-bridge chip you want to use. Most people don't run into this since they are usually controlling motors that are 7.2V, 9.6V, 12V, or even 24V so they are higher than the logic signals. If you are running the motors at a lover voltage or the same voltage as your logic then it can be a concern. I ran into this exact issue since I had a large motor that ran from a large 5V supply. The current requirements fit well within a pair of SN754410 chips stacked and since those run fine with a 5V supply for the motors that was the best choice.
Robert
Yes, the current would be controlled through PWM.
EDIT: Your power supply must output 8VDC minimum
If you should decide to use the L6205, I have attached the AN1762 (Application Note) for further reference information.
EDIT: For further information, click the "Design Support" tab resulting at the following link:
http://www.st.com/internet/analog/product/63232.jsp
Please note that according to the datasheet, the outputs of the L6205 can be paralleled for higher current.
@Loopy Byteloose
Have you tried to order samples? It sounds as though you did not even try. I would imagine they want these chips to be used everywhere in the world, but you may be right. I would at least give it a try and see what happens.
If your motors are running as 9.6V, 12V or more then this should work fine. With the 6V supply this chip really isn't going to work.
Robert
I really want to use the 6205, but right now im using a 9.6 nicad power supply, that doesnt leave much room to power the chip once the battery starts dying, and to boot your suppose to drain a ni cad all the way
I have several sample chips, but at this point I have not used them yet. If you intend to use these chips, I would definitely upgrade my power supply.
Bruce
Sorry, I did not try. Let's just say that it is personal choice as I don't require these in order to enjoy learning about microcontrollers. Many will ship the free samples, but want shipping. I have to pay via credit card and they may or may not accept credit cards. Often they want to ship their choice and I have different problems with different modes of shipping here.
I am simply a hobbyist. And at 64 years of age, I am not going to suddenly build a new business in electronics.
Having said all that, they seem fine to me, but others think the 8 volt minimum is a problem. Often, motors are operated at rather high DC voltages just to get the benefit of added power. Most of the world is not running on batteries. I see the PDF discusses operation at 48V supply. The higher voltage makes the h-bridge less wasteful in comparison to the voltage driving the motor.
I have noticed that an awful lot of users don't take into account that H-bridges require a voltage higher than the rating of motor. Even with MOSfets, there is some of this. You cannot just take a 6 volt supply and attach it to an h-bridge attached to a 6 volt motor and expect to get optimum power. Read the PDF carefully and accept the fact that you have to provide higher voltages to get a motor to run right.
Supposing that you upgrade the power supply, there are other ways to fully drain your batteries, such as another load. Just utilize the bateries until they no longer power the robot, switch batteries, and drain existing voltage with another load.
I agree completely. All of my stepper motors are rated for under 5V, but I run a 48VDC power supply to my drivers. My motors run great, much better than being supplied with a lower voltage.
So far the plan is to use 9.6v 2200ma/H nicad drill battery to run my whole system, which is 2 motors one parallax standard servo and a custom prop PCB with a switching regulation unit. I may be adding a stepper into the mix which needs 12 volts but that will be later.
These motors run ideally with a steady 6-7 volt power supply, they will run all the way down to 1.5v but thats just not enough power, 3vs is minimum. If I were to run the 6205, your saying I should definately upgrade my battery supply? 2200maH is a decent figure for motors using about .5amps if its not enough I can upgrade maybe for now i can just use 12 volt alkaline supply with 8 AA batterys.
As far as logic levels I need the 6205 to read 3.3vs and 5vs as Hi, so I can use prop/arduino/bs2. Now i have an extra 74lvc244 I can use for shifting logic from micros to the motor controller if thats the issue, but from what im reading its not. Somehow the logic level being higher than my lowest motor voltage is an issue?? The motor does fine at 6vs so I dont see why a 5v pwm signal would cause an issue?
Im pretty set as of now on the 6205, It seems to me like my system could handel it as is but would be better off with a 12v upgrade, correct?
You may just be better off getting something like this: http://www.pololu.com/catalog/product/1212
I don't mess around with standard DC motors that much, so I don't know what your motors can reasonably handle, but motors are often run much higher than their rated voltage. For instance, your motor might work perfect with a 24V niCad upgrade in combination with the L6205, but with the 9.6V, you are basically at the minimum.
As mentioned by RobotWorkshop, logic supply and motor supply are usually obtained from two different power supplies, but lower level voltages can be obtained from a higher voltage supply by using a switching regulator. For the sake of discussion, let's say that you upgrade to a 24VDC NiCad to power your motors, and you want to obtain a decent voltage for your control board, then you could use a premade module like this one (http://www.dimensionengineering.com/products/swadjhv), as suggested by Mike Green in this thread (http://forums.parallax.com/showthread.php?133037-Dear-Mr.-Parallax-High-Voltage-Propeller-Proto-Board), or you could create your own. My own preference is two voltage levels created from one power supply.
EDIT: I think you can also obtain a module from Peter Jakacki, as mentioned in the thread that you started, but I am not really sure if he sells them:
Ok I dont need to order a motor control module
So if I use a 6205 it would work like this correct..
I supply it with a pwm signal from a micro or 555 or whatever I want to the 6205's input pin
I then connect the 6205 to an 8 volt supply on its Vin pin
Now the 6205 will output 8 volts to my motor right assuming my Input pin recives a 100% duty cycle and the 6205 is perfect with no Vdrop, and if I set my PWM to 50% id have 4 volts to the motor (If this is wrong im sorry I really need some time to read the data sheet carefully, i just guessed this is how the chip would function seems logical)
My real confusion though is something robot workshop said. How does my motor voltage and logic level tie into each other, if the h-bridge is using its supply voltage to power the motor why would it matter to my motor what voltage my micro inputs to the h-bridge?
Now totally off topic but hopefully theres some insight here, I looked around wal mart today trying to find a light way to give my bot 12 volts, and decent amp hours I found these, bought two but plan to get 6 if it works out
6 of these babies in a parallel/series combo will give me 3.6 amp hours at 12volts, and there light as a feather!!! My question is can they be recharged a friend told me all lithium's are rechargeable and that these go in a camera that would usually recharge them, but idk if he was blowing smoke up my a**, thats why I only got two for now! I figue if he wasnt full of bs ill get four more anf find some kind of lithium charging IC at maxim or something. These babies are seriously the best power source ive ever seen 1.4 AMP hours and 6v a piece!!!! I looked at NiCADs and the 10 id need to make 12 volts is only 1.75 or 1.25 amp hours cant remember which, so im really hopping these lithiums work as there easily obtainable
The problem with too high a voltage to a motor is in the coil resistance as the combination of too much voltage and too many amps will overheat the coils and damage something. (But with the PWM, you are actually limiting that to a safe zone).
BTW, becaue they are MOSfet outputs, the L6205 can actually be used in parallel. Since one chip has two h-bridges, the logic and the outputs can be tied together to double the amps rating of the output.
Cool?
If I read the PDF correctly, the L6205 IC has internal voltage regulation that feeds the input logic the right levels for TTL control, and the final stage (the actual h-bridge) takes the higher voltage directly to the motor. This is NOT the case in other h-bridge designs. In those other h-bridges, one has to consider the i/o is right to get the final stage to perform correctly.
As previously mentioned, I am not an expert with standard DC motors, nor am I an expert on batteries. I initially mentioned the L6205, because I was attempting to solve an amperage problem, not a power supply problem. Additionally, I have no idea what kind of voltage your motors can handle and operate with in conjuction with PWM. However as I also previosly mentioned, many motors are powered with voltages much higher than their rating. The L6205 allows this to happen because it has current sensing resistors which set the maximum amount of current that ever reach the motor coils. Higher voltage power supplies are used to increase the speed to which maximum current is obtained.
One has nothing to do with the other.
I would recommend building a L6205 board, and test your motors with a variable power supply to see how it operates with different voltages. As long as you don't go hog wild with the voltage, I would imagine that you would have a wide range of acceptable voltage to power your motors.
What kind of motors are they? Do you have stats, datasheets, pictures, link to where they can be seen online?
It all depends on the context and the type of H-Bridge. For most of the ready made H-bridge chips those are two separate supplies and the motor voltage has nothing to do with the voltage on the logic side.
However, someone brought up building an H-bridge from scratch using MOSFET's and if you do that then it does matter what voltage you use on the logic side to control them. The normal logic levels (3.3v or 5V) won't be high enough to turn on the all the MOSFET's as needed. To turn on the high side MOSFET's you need to use a logic supply higher than the motor supply. Since your micro controller won't do this then you need an extra layer between them. This can be a special driver chip like the HIP4081 or you can build your own interfacing logic. By using an off the shelf H-bridge chip you avoid all the extra work since it is dealt with in the chip.
Robert
Id really like to try feeding them 12 volts but im kind of scared to do that, since I dont have any specs Im not sure what I would want to limit the current too if I feed them alot more voltage. The only specs I have currently are taken with my meter which is 2.5amp when motor i shaft is stuf and anywhere between .4 to 1.2ish amps while driving @ 6v.
If you are unaware of the amperage rating, it is better to error on the side of caution. Perhaps the L6205 is not the correct solution, or perhaps it might be. Either way, I certainly don't want you to blame me for release of magic smoke