Tronic's DIY dual Mosfet H-bridge (DSMhb)

Hi OscarH

Its been made this way cause the hardware PWM channels in AVR are limited. So I'd rather dedicate each axis a PWM signal as speed control and use the IN1 & IN2 as direction indicator.

And yes, if both IN1 & IN2 are HIGH or LOW then the motor brakes!!!!! Actually I use a small area near the setpoint position to brake the motor in order to be able to hold it in place if external force is applied and the motor has no wormgear.


Regards, Thanos


PS. The HPMhb-150A design is complete and I'm waiting the first prototype boards soon !!! :thbup:

Okay...the cat is out of the bag, so here is your teaser!

Presenting the HPMhb-150A...

R-eng

Hi Guys cant tell you how much im looking forward to building the new 150Amp H Bridge, any chance of a provisonal parts list, so that i can try to source the parts. I want to use this new Board to drive my electric wheelchair motors and then get my 2DOF FSX Flight simulator Motion Cockpit Running again, and seeing as how You guys here did such a great job on the first one i watching developments like a hawk. Cheers guys you have already got one more of many eager customers waiting to spend money LOL

Hi wannabeaflyer,

So far, I have sourced most of my parts from five different suppliers:

The boards from ExpressPCB (double-sided)
The IRFB3207 from Ebay seller.
The HIP4081AIPZ, LM35DT and ACS75xxxxx from Farnell
The optoisolators and regulators from Digikey
And the rest of small usual parts on my local electronics supplier...

Its not going to be an easy task for someone to purchase all these and surely not from a single supplier. We used parts that were absolutely necessary for the proper function and protection of both mosfets and motor in any case. It even includes an onboard ATtiny45 microcontroller that acts as watchdog, monitoring the current and temperature. We just wanted it to be as perfect as it gets!!

Complete part list is not available yet. And we have to build it and test it ourselfs first before presenting it here in the forum. A separate thread, dedicated to this new h-bridge, will be created. Please be patience!

Regards, Thanos

Hi Thanos 1 already Bought The IRFB3207 a while back when you 1st posted the ebay link ( took a gamble ) and will have to start looking for the other bits ... What was your experiance with Express PCb cost wise ?? my last PCB's were made by the Guy in germany great quailty but they were only single sided ... will let you know how i get on sourcing the HIP4081AIPZ, LM35DT and ACS75xxxxx ... my only problem with Farnell was their £30.00 min order charge but think this time i'll exceed that with the other bits so they may be the answer.. Saying thanks again for the work and effort you guys put into this we and a i mean the DIY simmers out there would be lost without the the sense of sharing that you and the other members so willingly :yippiee: give out Cheers :

Actually the prices are really good for double sided boards... If you live in USA !!! R-eng ordered them for me and achieved better price per board by ordering four of them. 25 euro per double sided board with with all holes plated through isn't bad at all!!!!!

The ACS75xxxxx is actually the ACS754 pachage CB-PFF.

R-eng choose ACS754KBC-150-PFF for 150A, and I choose the ACS754LCB-100-PFF as I don't intend to push any of my motors with more than 100A current!! The limits will be calculated as percent of the 0-5v output of the ACS754 so no firmware variations needed for different current sensors of this type.

http://www.allegromicro.com/en/Products ... /index.asp


Regards, Thanos

BTW- This is my design of the heatsink for the 16 MOSFET layout. It will be assembled from aluminum bar and angle of 0.125 in. Well balanced with a heatsink bank for every 2 MOSFET, lots of surface area, and lots of thermal mass. It will be cooled with a 92mm PC fan directly over them. With some clever tabs, one could even mount a 120mm fan.

All the parts can be sourced from Mouser and DigiKey in the US.

R-eng

PS. Perhaps a new thread is in order now?

Probably it is time for a HPMhb-150A thread soon :

Regards

Hey guys.. I am about to order the DSMhb and AMC1.5 - DSMhb1.2 from a PCB Maker here in Brazil. I gave up on trying to build them..too complex for me I would spend a lot of time learning that, i would get fired and no more $$ to proceed with my project..

Just to confirm, the latest version of the H-bridge is from tronic-s-diy-dual-mosfet-h-bridge-dsmhb-t820-50.html???
any special recomendations I should give to the pcb maker service provider??

Hi cauehawaii

You can use the layout I posted on the link below:
post7981.html#p7981

You can either give to the etching service the PDF file
DSMhb1_2b_sprint_bottom.pdf [61.68 KiB]

Or the gerber and excellon data file if they can use them:
DSMhb1_2b_gerber_excellon.zip [70.44 KiB]

There are no special instructions to be given. The layout is a simple single layer PCB (bottom side only) but in gerber data there is also an option to etch the few top side bridges and overlay the silkscreen parts description.




Regards, Thanos

Thanks a lot man.. there is no way back now.. I just hope this will not result on my wife giving me the divorce

The boards are ordered..aluminum bars ordered as well, I already have a universal joint and the seat. I am just missing the wiper motors but I will do some research on the list that Mambo provided me on my post (/post15006.html#p15006), perhaps I can find one cheap and reliable down here in Brazil.
Will I have problems with the boards if I use a different motor than wiper motors? I remember I saw some post about it somewhere in the forum..

thanks again for your help...I will be posting some more pics later next week..

Hi Cauehawaii,

Just remember some wipers motors have one side of the motor connected to chassis, and H-Bridges/HB25 need to have the 2 poles isolated for proper operation. This is also valid for any type of motor you may use.

OrscarH

I am actually thinking about buying some larger motors from Weg, I can get some used motors cheap here in Brazil.
I also have the option to use some pneumatic linear actuators, is it possible to use the actuators with these boards?

Hi Senetor,
This is with reference to Simple MOSFET Tester PDF. Just wondered whether you had built up and used used the circuit yourself, and if it worked OK. I juist did, but I got some really peculiar results. I first tested 15 'suspect' MOSFETs and all showed erratic results. I know for certain all of these can't be burnt out because they were changed as full sets as a precaytion. Then I tried to test some brand new IRF 3205s and these too showed something else.

If you post the URL, I can perhaps get back to the author and give him some details for a fix. I have been through the circuit several times, and changed both the 555's so there's no chance of any wiring error, etc. :?

Thanks, Erich

Hi Thanos Been awhile but have always been checking out developments here, and eargerly awaiting the next gen H Bridge, As you may remember i am still working on my 2DOF motion platform and recently won a bid for 3 Parvalux P3 LWS gearbox/Motor Combination units I Knew when i placed the bid that the Gearbox Output sped was Low at 8 RPM but the Torque was 51Nm so i figured i may be able to Change the Motor at some point and get better speed, My main question is i have the Controller recommeded By Ian at Budggies Built for fun but as money is getting tighter i would like to use Your H bridge PCBinstead of the MDO3 Motor controller Ian used.. I Have enough parts left over to Build another H Bridge and so Have one for Simforce GT Project and A seperate One for the Flight sim, As Ians System uses i2C protocol to drive the MD03 controllers i was wondering how and if i could Connect your H Bridge to do the same thing but without i2C.. is this Possible ?? i ask mainley becuse you did some testing and wondered if you went down the road of using your hardware instead of buying another Controller .. : By the way i am thinking of initally driving the motors at a higher voltage than the factory specs just to see if i can bump up the output speed before i dismantle the units :lol: will post the Pics when i get to the workshop .. (well it's a Shed Really but i have big visons for it ) Motor Specs 24VDC MOTOR GEARBOX MADE BY PARVALUX ENGLAND BOURNMOUTH

TYPE PM 3AWS/601254/3M

24VDC, 8 AMP, 120 WATT, RPM 2800 motor, RATED INT, INS CLASS F

GEARBOX Nm 51, RPM 8, RATIO 12/24

SHAFT DIA 15mm x 75mm LONG

Hi wannabeflyer.

You cannot use Ian's Picaxe controller with my DSMhb1.2b. As you said, Picaxe just sends I2C data signals to the MD03 that has no relation to the PWM pulse signals that DSMhb uses. The MD03 have onboard microcontroller that translates the I2C data signals to PWM signals for the h-bridge chip it has. Thats why MD03 is so expensive.

If you study more the way Picaxe works you will see that it doen't actually do many things apart from converting the serial data to I2C signals for the MD03 and send back to the PC the potentiometer position feedback data. All calculation are implemented in software!!!!

I'm not saying that Ian's way isn't good, in fact he made some cool addon's lately, but his electronics works only in conjunction with his BFF driver.

On the other side, I try to implement everything on hardware side keeping as simple is possible the serial comm part for speed and compatibility reasons.

BTW good luck with those motors. You need much more RPM output for a motion simulator. 40 - 60 PRM atleast.


Regards, Thanos

:hi: Thanos, as always was very glad to get your feedback on my request , understand what you explained so may have to spend the Money in the end .. i like Ians system for my Motion Cockpit because like your system it is relativily easy for the Common man to implement. I had to take a chance with those parvalux Motors even though i knew up front, that the speed was going to be a big issue .. i seem to recall in the past you advised that 40-60 RPM range was a good target and was thinking that if i Changed the input Motor for one with a faster speed then i may get what i need , or Drive the Motors @ 36-48 volts and use external gearing ( Chain/Sprocket ) to get the Optimum output speed ( But would loose some torque ) the reason for my way of thinking was that @ £13.00 each it was a massive saving cost wise instead of Paying £305.00 Each for these units New ( Downside would be reduction in unit life but may be a trade off worth taking if it gets my sim up and running ) ... the worst case is i'll mod one and if not succesfull then will put them back on Ebay to Recoup some of my expenses.. Actuators where can we DIY'ers ) Find cheap Actuators? : will keep you posted as to how this experiment works out either way. OH Oh while im here LOL:-0 What News h bridge 2 :thbup:

Hi Erichans,
I don't have a URL link for the Mosfet Tester, it was copied from a magazine called Silicon Chip. I have not built the circuit yet myself.
I have a link for another article on Mosfet testing.
http://www.4qdtec.com/mostest.html#simple
And this one about couterfeit components.
http://sound.westhost.com/counterfeit.htm
MOSFET failure mechanisms
http://www.4qd.co.uk/serv/MOSFETfail.html

Have you tried opening some of the cases to see what is inside.

Regards Senetor.

Hi Senetor,
And Hi too to all those who have, or are intending to, build the Simple Mosfet Tester. Please read this!
There is an error in the Circuit Diagram, hence all the weird results I was getting. The explanation part of the PDF page states that IC1, the first 555, is a 2 HZ astable oscillator, while the second is just an inverter---in short, when the output of IC1 switches LED1 on, LED2 will be off, and viceversa. In my case, with the circuit wired exactly as in 'Simple Mosfet Tester.pdf'', there was no alternate flicker of the LEDs at all. Either they were both on at the same time, off or some erratic in-between state. What I finally did, after a lot of checking and substituting (mucking about!!), was simple. Just reverse the polarities of LED2 and D2, leaving the polarities of LED1 and D1 as shown in the Diagram--that's it!!

While strictly not necessary, there are a couple of other things you can do. The green ON indicator light was far too bright. Increase it's 330 ohm resistor to 1 Kohm. LED1 and 2 are quite dim. Their 330 ohm resistors were reduced to 220 ohms.

The 4 tests applied by the Tester to 8xIRFZ44N's and 7xIRF4905, reveal that one each of IRFZ44N and IRF4905 is bad (Gate to Source short), and the 13 others are OK!! (One got lost!!). At least I can now put these onto the board with some measure of confidence! One gets pretty wasteful without a Tester!! The corrected Circuit Diagram is attached here.

Senetor, thanks for all that info you sent. 'Silicon Chip' is a good mag generally. They may have published a correction later.

Regards, Erich.

Hi Erichans,
Glad to see you got it working, the Mosfet Tester was posted in Circuit Notebook by a reader, not Silicon Chip itself.
Did you open any of the dead Mosfets yet, to see if they are the real deal and not counterfeits.
Also did you read the article on Mosfet Failure Mechanisms as it explains how brush and commutator noise can kill Mosfets.
It also talks of blocked motors and how disconecting a motor lead will blow mosfets, so do not put a fuse on the motor leads, just the H-Bridge power lead.

Updated Simple Mosfet Tester.pdf

Regards Senetor.