The fan I had was not within the operating range of a common drill battery. The EBM Papst fan was designed to work as a nominal 12v fan, but could safely operate from 8-14v DC. I needed from 16-20v DC. I didn't want to just plug the fan into a variable power supply and let her rip. I wanted to know if it would safely work at the desired voltage range. I took the fan apart and inspected the fan PCB. Again, this fan stunned me again and featured an open back design, not a sealed design. I could just monitor and probe the circuit components without disassembling or damaging the fan. I didn't need to crack open a rear fan housing or desolder the rotor from a PCB. It was all out in the open and ready for probing. The only drawback was the fan was upside down making the PCB hard to access. As any product developer knows, the solution is to just make a jig!
I constructed a sheet metal jig that held the fan in suspension so the PCB was facing upwards and not pressed against the back of a conductive metal shell. With a better view of the PCB, I took an inventory of the components on the motor drive board. The PWM and digital control of the fan was handled by a Microchip PIC16C62X micro, the 5V input source for the micro was handled by a ST 78L05a 5v regulator and the fets to drive the fan were Infineon 2n03l05 mosfets. These are all well documented components which was a godsend.
The good news is the ST 78L05a regulator can safely produce 5v output with as quoted a <=20v input. Since 78L05a regulators are made by several vendors, I checked out alternative datasheets. Some stated slightly above 20v as the highest extent of operation. The infineon 2n03l05 mosfet datasheet mentioned that 20v was within the range of normal operation, with minimal impact on performance. Other unforeseen problems with making a portable system based on a drill battery system was securing the legitimate battery connector. Many drill companies make sourcing their proprietary connectors or connector assemblies impossible. luckily, I came across the Dewalt genuine 20v port part after searching through an older portable vacuum unit repair page. It is available directly from Dewalt service net and also via 3rd party suppliers. The connector is not exactly cheap, but it is a legitimate connector that does not rely on using 1/4in spade connectors shoved in a plastic housing with a little added super glue for good luck.
At this point I'm still on the hunt for the perfect potentiometer. I was not anticipating the potentiometer search to be so challenging. It turns out that there are an incredible amount of variety with panel mount potentiometers, as long as you are looking for one that does not incorporate an on/off switch. I am after a linear potentiometer instead of a logarithmic one as linear should offer the most predictable gradual control best suited for this application. A quick Digikey search reveals 698 types of linear panel mount pots with solder lugs and only 22 that include a single pole, single throw switch (SPST). None of those 22 are fully enclosed potentiometers either.
Unless I find a better SPST linear potentiometer, I may decide to cave in and settle with a logarithmic one or move the on/off button to potentially the handle. The handle option would definitely add to the cost and complexity of the unit because there is no easy way of mounting a button board inside the handle. A separate PCB, tiny metal bracket and drilling jig would need to be set up to achieve a handle mount button. However, if that means having a nicer potentiometer, it may just be worth the effort.