HPMhb-150A (high power h-bridge)

So here is the latest teaser.

This is an initial run of the HPMhb-150A with 12VDC batt input running a wiper motor. The thermal and current overload features are not yet active (they will require firmware for the ATtiny mcu), but the optocouplers to separate logic from power circuits are fully implemented. This test was a preliminary one where only 1 MOSFET per quadrant is used. The full compliment will be 4 per quadrant for a total of 16. Rated at 75A each, that would be a max continuous of 300A with proper heat sinking and fan (already fully designed). The heatsinks on this test are just basic ones. It is conservatively rated at 150A, and that is the limit of the current sensor used.

I just tested with dual-batteries at 24V, and the input can be up to 48V nominal, allowing for the slight over voltage of fully charged batteries.

You can see it in operation here...

http://s529.photobucket.com/albums/dd33 ... yrun02.flv

R-eng

Hi,

Nice to see it run so smoothly!!!! I think you can notice the difference in the status leds if you compare with the older DSMhb1.2b... The HIP chip really does what is designed to do!!!


R-eng, edit your video link above. I had to make it copy paste for the link to work.
http://s529.photobucket.com/albums/...action=view&current=HPMhb-150A2nddayrun02.flv


Regards, Thanos

I know some people have mentioned them around here before, but is this board much better than the OSMC board ? ( or are they completely different products )

thanks
T

Hi,

The OSMC motor controller is similar but it miss a lot of features that HPMhb-150A has...

Let me name few:

-Optocouplers to isolate completely the logic from the power part,
-Single PWM input (speed) with two digital direction controls, to be fully compatible with AMC motion controller, and ability to daisy-chain two HPMhb-150A boards through the same control cable.
-Watchdog microcontroller that monitors current and temperature level to automatically shut down the h-bridge in case of emergency, saving the mosfets.
-Possibility for I2C communication between AMC and the watchdog microcontroller to set the current and temperature level limits.

Thanos

First off, let me give credit where credit is due.

The OSMC board design is the basis from which I began the design of the HPMhb-150A. That H-bridge is a tremendous development of simplicity and size to output capability. It has consistently proven itself in robotics.

But, after burning through numerous MOSFET trying to drive high powered motors with the excellent DSMhb, it became obvious that not only a more powerful H-bridge was needed, such as the OSMC project, but one that had generous and multiple protection features built in!

So began the HPMhb-150A and Thanos' generous help with the design mission. Thanos has just stated the main differences, but there are a few more such as...

-thermal and current overload components built in
-slightly larger size to allow for much larger heatsinking
-better connection facility using bolts/lugs for higher current capacity
-onboard main fuse protection
-additional 5V supply (besides 12V) dedicated to the power section
-much thorough power supply filtering for 5V, 12V, and Batt V inputs
-more flexibility using jumpers in how the power supplies are distributed
-very large planes for the power supplies and motor drive lines (no need for trace buildup with solder).
-included more TVS (Transient Surge Suppresion) components
-motor drive LEDs which can be remotely mounted to enclosure facia

I wish to thank all those who participated in the design of the OSMC, for without them, this project might not have developed quite so fast and efficiently (i.e. less mistakes).

But there is more work to be done to finish the project, with thorough testing before release.

R-eng

Hi Guys iM at crucial Phase with My developments, i Noticed that you indicated that the new H-Bridge will be able to Communicate using I2C :clap: Can i ask and No Offence intended, its just that with my 2 DOF Motion Platform i Have a number of Options as to how to Drive the actuators and get Position Feedback data from sensors ..Some of you guys may know the work Done by Ian at buggiesBuilt4fun and as i have his software and some of the hardware i really dont want to duplicate Expensive Motor Controllers : Ian Uses MDO3 controller made by Devantech which communicate via I2C but are only rated at 25Amps . so if i could use the new high Power H-Bridge as a Replacement for This Controller i would Be in a Great Position to have the flexibility to use various combinations for my Projects .. so Guys would it be possible to use the new H-Bridge with Ians Software for My Flight Sim Platform : Had to ask guys dont shoot me down ..Money is tight and lots of projects started that i would like to finish and make avialavle here at last done enough lurking wanna get moving :happy:

I think I explained with the phrase: Possibility for I2C communication between AMC and the watchdog microcontroller to set the current and temperature level limits.

It will save your time from removing the Attiny45 chip to reprogram it with other Temperature/Current limits...!!!! Its just a stand alone watchdog protection circuit!!!!!!! :lol:

Ian's software can talk through serial protocol to my AMC1.5 and then the AMC1.5 can move the HPMhb-150A as simple as that... The HPMhb-150A cannot be moved by the I2C signals that Ian's Picaxe hardware have...


The High Power H-Bridge is not just a Replacement for the Devantech MD03... Its a total different thing and can't be compared, in price and performance... :mug1:



Regards, Thanos

Nice work.
I want ask, why it could handle only 48 V ? Is it because type of transistors ? I ask because many servos (from old cnc) have higher voltage.

The design criteria and the need to convert the voltage from the battery(ies) 12-60VDC to12VDC with a wide range and dependable DC-DC converter (so the HIP driver can be used), predicated the limitation to 48V nominal. The component selection fulfills the additional headroom needed to handle the slightly higher voltages present in freshly charged batteries when 4 are used in series, without bumping up to the ragged edge!

Seriously, how many need to have more than 48V nominal??? If that is the case, then you need to look into industrial solutions and their higher $$$.

And aren't old CNC servos industrial items?

P.S. another distinction from the OSMC is that it could only handle 50V max, limiting your batteries to a safe 36V nominal, based on the design/recommended BOM.

R-eng

Cheers Thanos and everyone involved in the development of the HPMhb Project . Had to ask the quesion because for me money is getting tight and having started these projects i just want to get on with and start to enjoy them instead of constantly tweaking and waiting for ways to make them better ( a lot of Modding not enough driving/flying) . so just to say thanks for the heads up. will now have a look at all the interconnection options and finally decide which way to go. At this stage becuase of my investment on Both Ian's Method and X-sim/Thanos system. i need to save money and maximise the equipment i have already got, hence the need to see how the new eagerly awaited HPMhb was coming on, afterall i would rather put my money on the HPMhb as a member of this group greta work as allways guys. :

For example this
I am not electronic guy so i am not sure of benefits using higher voltage but i thought that motor have better dynamic characteristics with higher voltage and if you use transformator it could be chepar (due to less current).

So main reason is internal power suply for electronic .

I am not complain, i am only curious.

Thanks to Thanos and r-eng for the comprehensive replies. And thanks also for your work developing this new h-bridge

lookling forward to it !
t

@ Mambo

How is the cost of a suitable transformer with multiple voltage taps, and other associated electrical/electronic parts, going to make it cheaper? And you would still need the entire HPMhb-150A (or DSMhb) to move the simulator.

Why not try this motor if you need 1000W...

http://tncscooters.com/product.php?sku=106175

R-eng

Less amps -> smaller diameter of wire, bigger voltage (smaller ratio) -> less winding. So ideal voltage would be around 230 so we would need no transformator .
Why they do more types of same motor ? What i know from stepper motor is if you want better characteristic from it you must use higher voltage. I dont know if its same with DC motors .

Great job on the H-bridge controller! Looking forward to further updates! Any chance for a schematic? Which IC are you using? I have been looking into this project with a hip4082 controller. Are you guys using the Attiny only for temp and current monitoring? If so, this protection feature can be done a lot simpler and cheaper with a dual comparator for ex.
Implementing a microcontroller with current sensors give a lot of headroom to fill with different h-bridge driving modes described here: http://www.modularcircuits.com/h-bridge_secrets2.htm and also protective features. If I was to use a separate microcontroller, I would definietly not waste it for simple monitoring and 1bit decision making. What's your thoughts?

Mambo:
Its not so simple. Driving stepper motors, AC motors and DC motors differ in a lot of aspects. You can't just have a lot of Volts and assume you will have a lot less Amps. In fact, its the opposite. Just try licking the terminals of a 9V battery and a 24V battery... ok, maybe don't try the 24V because you might hurt yourself. 230V(330Vp) will fry about 97% of the semiconductor parts on the market, so the H-bridge controller would go up in smoke as soon as you hook it up. Not making it anymore complicated, high voltage is not the way to go for our purpose.

www.allaboutcircuits.com has a lot of great articles about motors, electromagnetism, and everything else electrical/electronic, highly recommended!

Good to know.

We decided to use a really small microcontroller (ATtiny45) for the watchdog function as this gives us the opportunity to program ourselves the limits of each sensor, and the windows in which will be active. Hell, we also can add shutdown delay timer in case of overheating in slow rate indicating that the extra heat is no real problem but a temporary load (like the extra bumping motion you get when you get off the track for a while).

And the best part with using microcontroller is that you can declare these limits with real numbers defined in Ampere or degrees of Celsius. Nothing simpler than that for the average simmer, that don't know how to calibrate the old style comparators with multimeter and advanced math equations...


Regards, Thanos

Thanos, I agree that it would be very nice and I would prefer the microcontroller as well for a fully programable and adjustable control system. All I'm saying is that it could do so much more with a bit of programming once the communication is working.

Using a comparator wouldn't be any much harder in my opinion, the only difference is that the temp and current limits would be set by 2 calibrated pots on the controller. There would be given min and max values on the pots and they would adjust proportionally in between, fairly straight forward. A delay is also possible, in fact a thermal delay is already there if you put the sensor further away from the fets. It would definetly be cheaper, less complicated and just as effective until the Attiny is working completely. I will go this way and let you guys know and decide what is easier and better to implement.

The ATtiny is a functional part of the HPMhb-150A pcb already. The entire circuitry has been designed for awhile now and there is no anticipation for any changes, other than very small value changes for some of the passive parts so all functions operate at max efficiency.

The current and thermal limits, as well as thermal recovery is set by defined values, which are clearly preferred to some arbitrary position on a pot. If I want the current to shut off at 57A or 103A or whatever, up to 150A, then that is what I get. No guessing, no approximations, no ambiguity. Same with the temp, who's delay is then set by a thermal drop from the set point, not some ill defined time delay. All this is easily and precisely set by the user, for their own simulator demands and limits, individually for each motor. That is precision and efficiency of protection. This was the design criteria for the HPMhb-150A. It is very simple and inexpensive.

BTW, knowing and setting the temp precisely at the point of desire, the MOSFET, is the preferred way here. And not using it in some variable delay scheme with distance manipulation. How on earth would you determine and control all the thermal variables possible with even a small remote location?

The method used is pretty inexpensive, is not complicated, and is extremely effective. Done!

R-eng