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

Hi Senetor,
I didn't need to open any of the MOSFETs. They were working perfectly and I'm sure I caused the problem----poor fan and no fuses, long untwisted motor leads and no noise suppression capacitors across the brushes, 1 MFD/600V---all rectified now (see Page 13 of this thread)! 4QD is a great site, but I did disobey one rule. I added an ammeter in the line! R-eng had suggested using a clamp meter, but that's OK for temporary testing. I took a bit of courage from Thanos who had the inductive one on line on his Joyrider.

The Tester has acquitted 13 of 15 'suspects'. For a second opinion I'll also rig up the one 4QD has given. Anyway, I'm building a second DSMhb as backup with copper heatsinks and powerful fan for testing a new ~100W motor I've made up. (details will be here after testing) Pity I had only the 'no TC' board handy so I'm using up that.

One thing is certain. We need a good MOSFET tester, else your museum gets filled up fast, even with good ones!

Regards, Erich.

@ Erich,

I built this circuit about a month ago and also encountered problems. I was thinking of replacing the 555 chips when today I saw your solution. I rotated the LED and diode and everything worked perfectly...a big thanks!

The explanation of the LED sequence is a bit untidy, so here is what may be a clearer one.

BTW, I am methodical bench testing the prototype of the HPMhb-150A at this time. There are some hiccups, so it will take a bit more time before it powers up some motors. But it's not far off!

R-eng

Regarding the resistors for the LED's, it really depends on their mcd outputs. The original values proved perfect for me. Experimentation between 220 and 1k is needed to normalize the outputs.

Here is my board.

R-eng

Hey, nice little board! Any PCB layout of it that we can use to make it? :clap:

I have all the parts, but was too busy to actual make the board on protoboard...


Regards, Thanos

Dual sided expresspcb layout.

R-eng

Hi Thanos just wanted to ask your advice again regarding my Motors and the Spped issue.. have designed a Geared pulley driven system which will be linked to my Motion platform using 6mm Steel cable, becuse my Motor/Gearbox combination is so slow i have thought about using 3/8 pitch chain and sprokets to get the output spped up to 64RPM ( well in theory at least :lol: ) all the main element will rotate in bearings and the Pulleys will be 100-150mm diameter . have a look at the cad drawings i did to see what you think of this option .. if you guys can see any problems with this method i would appreciate tihe feedback before i start making the bits ... on the question of out put torque i had to loose a lot to get the speed but am waiting for a answer from parvalux about how easy it may be to change the Motor to PM60 which spins at a faster speed so i wont have to double up the supply voltage , or cause the unit to wear out quicker by my dodgy calculations i figure that with an input speed of 8 rpm @24V i should get 64 RPM with the gearing set at 8:1 the torrque of 51Nm will be in excess of 500Nm at arround 100mm from shaft which would give me +/- 200mm of cable travel and easily allow me to achieve 15-20Degrees Pitch and roll ??? What do you guys think.. Thanos if this is in the wrong place i'll move it to Actuator designs section : sorry about the large second PDF could not get infraview to make it smaller :-(

Hi,
As a follow up to the Simple Mosfet Tester fix on this page, and for the sake of completeness, I tabled the 4 tests from the text in the original PDF. There are two versions here, one for those who will/or have used LEDs exactly as per the circuit, or for those who might have used green and red LEDs for LED1 and LED2 respectively.

R-eng, thanks for checking these out.


Regards, Erich

Hi Rene,
Good idea! May I suggest however, that we wait awhile until MOSFETS that read OK actually work in-circuit. At least 3 of us have it now and when someone gives a :thbup: maybe then we can proceed. I have a mixture of 13 and am just building a second DSMhb, so I'll feedback when I have a result. As for everybody, time is my huge constraint!!

As I say, just a suggestion!

Regards, Erich.

Hi,

The way you have designed, it might work. But I don't know if the steel cable will leave you with much movement of the output shaft, as the tension fittings will hit each pulley disk, if you don't limit the rotation. I'd suggest to measure the length of the pulley cable and use an engine belt near to your size calculations. If not possible to find exact size, make the small pulley adjastable to tension it later.

Regards, Thanos

Just an update i am planning to buy pulley wheels with size circumferance which is going to be equal to or bigger than the required travel and would then allow for 4/5th of the circumferance to hold the cable ... by this is mean that at least half a turn each pulley would wind/unwind at least 200mm of 6mm diameter cable ( ONe pully for each cable so they cant get tangled ) : Thanks again Thanos

Hi Rene,
Nice to hear---great project!!

Some quick advice before you get in too deep---order the DSMhb PCB as the 'TC' (terminal connector) design. Easier to fit and/or replace MOSFETs. The next one may be a bit of overkill, but I'm using copper heatsinks for the second DSMhb. Also, I replaced the plastic spacers between the pairs by copper tubes. I changed the 75mm rotor fan to an 85mm more powerful one. I used card deflectors to better direct the airflow over the heatsinks. and last, I put the whole lot in a black box with the LCD, reset switch and serial ports sticking out of the box for easy access. There are 2 20x60mm exhaust ports at the bottom of the box. I could not get very good quality sockets for the BC546s and the crystal. So now I've soldered the transistors in, and after finding the crystal that worked stably, soldered that in too. But if you get nice tight sockets--OK!!

Thanos held my hand throughout all my project problems (see Pp 31 and 32 (under that hardware thread). Also Page 39. And lots more around about there.

Good luck with the project, and hope the 24 MHz crystal works!!! :thbup:

Regards, Erich.

The 24MHz crystal does work, but I understand from Thanos that with the new firmware, the 22MHz is the ideal version.

R-eng

Hi R-eng,
Am I missing something here? Which is the 'new' firmware to which you refer?

Thank for the pointer, and Regards,

Erich.

Hi Erich,

The new firmware R-eng is taking about is the v1.1. It implements some fixes on the PWM curve near zero value, and some more fixes on the LCD update code. I have not yet released this new firmware cause I was hopping to discard the Hyperterminal setting method and have an external GUI application (written in VB) to be able to change a lot more parameters and hopefully use some profiles that can be shared. Still need help with it so if anyone willing to help me with VB is welcome...

In meantime I already have ready a 3DOF firmware (probably will call it 2.0) that is used in combined axis motion platforms (like EvanF's chair of torture if you remember it). I'm trying to develop additional code for it so it can use quadrature optical encoders for position feedback. I found a really cheap solution in quadrature encoding that will present to the public soon I hope. Just imagine paying $15 for an 128CPR encoder with 16-bit onboard decoder microcontroller, and best of all, ready to buy off the shelve...!!!



I believe I now have more projects in progress than ever!!!! I don't even dare to number them :lol:


Regards, Thanos

Hi Thanos,
Great to hear that you are still developing the firmware minus HT and with profiling features :clap:
Look forward to using with my second AMC-DSMhb set for my second smaller, but 3 DOF sim (with manual yaw!!! LOL! in my mind at this moment). It will have more powerful motors with the WW motor gearboxes only grafted on, and the plastic pinions replaced with steel ones. I hope I'll be able to get hold of the encoders you mention, as I'm still struggling to get SS495A's.

My grandson gets #1 Sim---but he doesn't know it yet! ;D

All the best with your great projects!!

Regards,

Erich.

Hi,

I have build a new h-bridge...

My next problem is that the left motor (motor2) starts shaking when it is in middle position
and the status leds are flickering (also with no motor connected).

Motor1 is working fine.

I swapped motors... same thing. Think it is the signal from motor2. I also re-programmed
the atmega and tried different kp-settings. No succes...

When I start the profiler only 1 axis is showed and both motors move. Both in different directions.

Does anybody have an idea of what this might be?

gr. Jeroen

Hi,

When I turn the pot which is connected to motor1 the leds smoothly go on and of like they should...
But the leds of motor2 are switching on off (not smooth from green to red or reversed).
The pot of motor2 is working fine on the lcd.

gr. Jeroen