CF-18 mk1 testing thread

I'll start 'my own thread', if that's allowed, to store and also share all the findings of the tests I've done to a bunch of CF-18 mk1:s I got couple of weeks ago.

The story so far...

Still figuring out how best to integrate GPS and serial port from the schematics (the thread on requesting schematics)

My first test where I froze the toughbook to -28C was to find out how the HD heater worked and if disabling it would save significant battery power. Conclusion: it doesn't (less than 10%).

Today I tested the car adaptor CF-LND1224A (Automobile Adapter Model
, Manuf: Lind). Also got some idea of the different power consumptions of various features:

(units in amps drawn from battery, adaptor should be at least 90% efficient, battery voltage under load about 12V):
Start and boot: 2.3 max peak
Windows loading: 1.5 - 1.8
Normal use: 1.5 - 1.8
Battery unplugged: 1.0

Screen brightness: +0.3
Battery loading: +0.5 - +0.8
HD activity: +0.2

Powering 'OFF' WLAN, Bluetoothin, or WWAN have no significant effect on the power consumption.

Switching transients go through the LIND Auto-adapter and trigger the power-switch of the CF-18: turning it on or out of suspend mode or putting it back to suspend (if that's in the windows settings.). Need to buy some toroids and hope that helps.

 

+Rep for your time & effort.

Out of interest, a question about the GPS systems that include an integrated bluetooth.

When you Power Off, if the battery is still attached is there any draw on the battery?

o.g.

 

My CF-18 doesn't yet have a that module but I'm thinking of modifying it to use a Fastrax uPatch 100 module via the infrared-serial port.

Generally the CF-18 does have power pins around the motherboard that are supplied with +5 even when the power is off - for the purpose of battery backed-up power. Anything that is connected to such pins is supposed to draw an insignificant amount of power so as not to drain the battery too quicly in storage.

My module for example uses a whopping 44mA of 5V (0.22W) when active but only 150uA (0.00075W) when in battery backup mode (draining the battery in about 7 years). The battery input for the module only supplies the GPS memory and RTC (Real-Time Clock) so that the module gets a quicker hold of the fix when it wakes up next time.

 

My original battery was giving me less than 1 hour of power so I decided to refill it. Opening the battery case requires some doing - I used a heavy tool-knife to pry open the back end (opposite end to the connector) by placing the sharp blade in the seam and using a twisting motion. Ended up breaking most of the plastic 'fingers' holding the back together. Nothing that a bit of glue, tape or epoxy won't fix later. Also the backend is safe to work with since there's less danger of shorting out anything important - no wires and the cell-ends facing there are the same polarity.




For the experiment I bought two 2x3 cell battery packs ('Wamtechnik' CGR 18650+PCB). The packs contain the original japanese cells and interestingly a control circuitboard plus an inline fuse (both of which I discarded) - datasheet here: AYAA TECHNOLOGY CO.,LTD - Products - PCM-Li01S8-015/PCM-CS01SH-22 (with NTC) ) The cells have tabs welded so no need to solder anything to cells surfaces. Two packs cost 63 eur.




The rest is just soldering. I needed to add an extra wire (black one) where the old minus tab was directly soldered to the control board (red=plus, blue middle, black=minus).



Ready to be closed with some insulating and protective pieces of tape over the soldered connections.

The battery now requires a battery refresh procedure found in the machine BIOS (F2 and start up, the battery refresh option is found in the 'Exit' menu of BIOS)

After the procedure, windows battery meter indicated that there was '6:23 hours left'. With all power-saving turned off (except display which was in low setting) the battery lasted 2:45 under simulated use (constant 25% CPU load).

Next have to test if doing another battery refresh might give even more battery capacity?

 

Yeah, you're not likely going to ever get an accurate battery meter with that rebuilt pack; the control PCB has a chip which will "remember" the discharge curve of the old cells even after they've been replaced; it is supposedly possible to "reset" that . You may well find that half of your battery life is after the battery reads "0%" and this condition never goes away; if you shut it off or hibernate in that state, it won't turn back on.

We've been beating our heads against this issue for years on this forum; do a search for "FOAD chip" & "FOAD timer" and you'll see what I mean.

mnem
Dumb looks are still free.