Continuing education

MCU projects

Amateur radio

Broadcast radio

High voltage



Experiments with DeWalt DW954 cordless drill

I was given it as a Christmas gift in 2000 and it proved to be a good solid tool for chores around the house. Everything worked well until one day when it failed drilling a teensy-wheensy hole in 2x4". What! I was quite a bit surprised because it wasn't even metal. A little puff of smoke came out of the motor compartment and that was it. By that time I had grown to love it and didn't want to throw it away. So it seemed only natural to look into the belly of the beast. After having gutted out the control board I noticed that the power MOSFET as well as the freewheeling diode had departed this world. Their legs were soldered to the control board and they themselves were bolted to the outside of the trigger assembly that doubled as heat sink. Replacing them seemed straightforward. But replacing with what? Both had only a few letters stenciled on them and no other markings. The control board had only one IC with "555" on it. The "555" and a cluster of SMD components around it seemed to be consistent with a 555 timer-based simple PWM circuit. I am fairly certain that at the heart of the control circuit lies something like this.

A little bit of "dead travel" is mechanically built into the trigger movement and the motor gets no power at first. Once past the threshold, it starts moving the slider and the PWM starts cranking out short pulses at approx. 400Hz. As the trigger is depressed more, the duty cycle increases - thus increasing power fed to the motor. Once it was clear how everything worked it was time to replace the burnt out component. But replace them with what for both were only scarcely marked? After a little bit of looking around I found someone's thread on Edaboard, asking similar questions. They recommended the IRF305 MOSFET. I didn't have it but a quick search in my junk box found an IRF1010 (Vdss=60V, Rds=12mo and Id=84A). However, IRF1010 had a quite substantial input capacitance, around 10,000pF according to the datasheet. I wasn't sure if the CMOS timer could drive that. Also, IRF1010 wasn't a logic-level device. That concern was put to rest after observing timer's output with an oscilloscope. It showed output pulses from 0V all the way up to VDD so the logic-level device wasn't a necessety here. To make the long story short I ended up using what I had and an IRF's 80SQ045 (Ifav=8A, Vrrm=45V) freewheeling diode. I guess the relatively low switching frequency worked in my favor, too.

Now the next question rose. I had a pair of old Thunderpower 2050mAH 12V Li-Po batteries donated by my friend Tony. They don't pack enough punch for flying anymore where you have to get the whole charge out in the hurry, resulting in 20-25A currents. But for powering a drill? Could I use those in series instead of the native 14-cell Ni-MH pack? If the power supply was 24V then the MOSFET would have to withstand at least that. The same goes for the diode. Luckily, according to the datasheets, both were OK. Of course, being that the Li-Po's were quite a bit longer, I would have to create a new housing for them.

I ended up using a "container" that is permanently mouted under the handle. My container came to measure 6" long, 1.75" tall and 3" wide. That gave me plenty of room inside to house my super-duper batteries. As sides, I used 1/8" playwood that I had at hand. Both the top and bottom of the housing I cut out of 1/8" masonite that came from junked radio's back wall. As you can see in the picture below, there are silkscreened letters "FM ANTENNA" still visible.

Here you can see the completed Li-Po battery housing. Of course, my housing looks kind of boxy, unlike the the original battery shape. But the unintended benefit is, it's much more stable than the original, owing it to much larger surface area.

Here is a shot with the bottom cover removed. To make sure the bottom cover always goes on in the same orientation, I have marked the frontal direction inside.

Sometimes after drilling a hole you put the tool away without thinking and it gets bumped. These batteries are pretty fragile, they are kind of bags. Therefore care must be taken so that they are protected from banging around inside the box. That what the pink foam is for. The foam also acts as a heat trap but that aspect so far has not bothered me. I guess it depends on what you use the drill for. If used for really heavy duty jobs then ventilation holes should be drilled in the housing.

That's it! The job is finished. I have been using my DeWalt like this since Nov 2008 and I love it. Unlike native Ni-MH's, Li-Po batteries do not have significant self-discharge. You can leave the charged tool alone for months and when needed, it purrs into life without hesitation.


One thing needs to be stressed extra.

Li-Po batteries can only be recharged with a special charger that is specifically meant for them. Any other recharging practice will result in destruction of batteries and may result in fire as well.