January 2001

Tuesday 2nd January 2001

With the recent death of yet another DC DC converter, and the ongoing timing problems I was having with two MOSFET being on at the same time and shorting out the power rails, I did something I should have done months ago. I had a really close look at the MOSFET's data sheet! In particular I looked at the input gate capacitance values and compared these with those from the lower powered MOSFETs I had first used before being given the high powered ones from Fairchild semiconductors. I noticed that the capacitance values were about twice the size, and the threshold gate voltage was lower. This combined with the fact that I was now connecting three MOSFETs in parallel meant that the input gate capacitance was now quite significant. On top of this, my initial design for driving the gate voltage off was by using a pull down resistor. After going over the maths of this circuit, I calculated that in the worst case, it might take nearly 1 millisecond for the MOSFET to switch off. This would explain why the MOSFETs were shorting out the rails. I think I've learnt my first lesson; "ALWAYS READ THE DATA SHEET IN DETAIL". The only sensible option I can think of now is to do a major rework of the circuit by including a lot more suppression and de-coupling of the power rails, together with a "push-pull" driver stage to each branch of the MOSFETs "H bridge" circuit. Watch out RS, here comes another order!


Go to top top

Monday 8 th January 2001

I contacted Fairchild and begged for some more MOSFETs so that I had enough to complete the driver stage for both motor circuits. I had calculated that I needed at least four MOSFETs per branch to handle the 500 amp stall current. Tim from Fairchild obligingly said some more transistors were in the post, and offered some useful advice on the redesign. He also said that if I wanted to send him a copy of my new design, he would have a look at it, and make any recommendations that he thought might improve it. Well, that was an offer I couldn't turn down. He also pointed me towards "On Semi" who are a semiconductor company with a very informative web site. After a brief surf of their site I found a couple of amazingly pertinent documents outlining the various considerations you need to take into account when designing an H bridge motor dive circuit. Armed with this info I spent the evening redrawing my circuit.


Go to toptop

Tuesday 9 th January 2001

I sent my new motor drive circuit off to Tim at Fairchild for his comments. Hopefully they won't be too critical. I had a chat with Bill from Settform, a sheet metal work shop that was recommended to me by Ian my usual metal supplier. Bill was pleased to help when I explained I was building a robot for Robot wars, so hopefully he won't charge me an arm and a leg for the jobs. The job we discussed today was to mill a long key way slot into each of the wheel axle shafts. Other jobs I mentioned were to make the idler shafts for the motors, and to build an aluminium block assembly to mount the idler shaft and bearings on. The motor job would be done at a later stage when I had the bearings, but Bill said they were happy to do the work, and would price it sensibly.
I contacted SKF to ask if they would be willing to donate the additional bearings I needed for the motors. Steve wasn't there, but I left a message hoping that they would return my call with a positive answer. Phoned Steve again later and he confirmed SKF would be happy to donate the bearings. It must be my lucky day.


Go to toptop

Wednesday 17 th January 2001

I had a reply back from Tim at Fairchild saying that they had had a look at my new circuit design, and it looked okay. Tim also said that the MOSFETs were being despatched from Korea, so to expect a weeks delay in delivery.


Go to toptop

Thursday 25th January 2001

Another 20 free MOSFETS arrived from Fairchild so now I have enough transistors to complete the two power stage. All I need now is the time to build it!


Go to toptop
 

Friday 26th January 2001

Eight second hand go kart tyres arrived today from Andrew of the Tornado team. I now have some contingency if I get a flat (or two, or three, or four...).

Go to toptop
     
Go to top top

Back to Hassocks Hog home pageBackward to December 2000Forward to February 2001

Last updated 19th Feb