Penny & Giles TrackerMouse

Archie
Site Admin
Posts: 113
Joined: 31 Jan 2019 15:27

Penny & Giles TrackerMouse

Post by Archie » 03 May 2019 13:07

Ball size: 11/2" (37.64mm)
Interface: Infrared proprietary
Year: 1984

Image


Quite advanced trackball, created back in the middle of '80s. As far as I know it was the first trackball specifically developed for portable personal computers, and the first wireless trackball in history. Maybe it's also the first combo device with possibility to use it as a mouse, and probably the only one using optional fiber optic cable connection. It's clear the creators wanted to make the device as versatile as possible.

Version I've got was made for ACT Apricot machines. My understanding is, the device was initially manufactured for them per Applied Computer Techniques order. Later P&G marketed it under its own name as TrackerMouse and TrackerMouse Plus, and different variants were issued: with quadrature, PS/2, serial and/or infrared interfaces, with added solar-powered calculator, and even with keypad for macro launching, taking almost all of its top cover space. Most of them were ivory, except for black model for Apricot Portable.

Image Image

Image Image


Unusual design with ball protruding also from bottom, have explanation: an option to use this device as a mouse, moving it on the desk surface. It was noted in the Apricot manuals and ads: "Infra-red Mouse The Mouse for the Portable is identical to the Mouse for the F1 (apart from the colour of the plastics). It has been designed to be used either as a Mouse (by tipping it forward and rolling it along the desk), or as a tracker ball (keeping the Mouse stationary and moving the ball by finger movements)". Additionally it's possible to operate the ball with thumb, holding the device in hand, and optionally supporting it by finger from the other side - a bit similar to the famous Logitech Marble FX design.

From my POW, the "mouse mode" is essentially unusable - but maybe someone actually wanted to use it that way. Anyhow, near the end of device's production (seemingly ended in 1990) it was advertised purely as a trackball, with no any mention of "mouse" capability at all. As a trackball, the device is very impressive. Its flat shape looks "too simple" - but actual use is really convenient. Long narrow buttons are equally accessible in any hand position. Ball movement is very light and smooth. My only objection is a stiff button switches: they're simple tact buttons modified for side mounting, relatively hard to click. In some of late reviews microswitches were mentioned, so probably it was improved after first batches.

1.5" ball is supported by ruby gliding points, and they're not usual spherical ones, but merely discs. 3 of them are glued to the bronze plate acting as a base for whole assembly, and 3 more glued to the inside of retaining ring, coming into action if mouse mode is used:

Image Image


Sensor discs are of radially striped reflective type, with pair of Optec OPB706 sensors looking at them. Discs attached to the bronze shafts rotate in high-grade micro flanged ball bearings (two per axis); protruding parts contacting with the ball are covered with rubber-like tubes. Each sensor mechanism is mounted on pivot block, and supported by leaf spring. Quality of mechanical components is very high. Calculated resolution is 410 PPR / 88 CPI.

Image Image

Image Image

Image Image


Being wireless, the device is powered by 6-volt alkaline battery. In first model, it was 4LR61 a.k.a. "J-type" one, extremely rare now - and it was a big challenge for me to find this battery nowadays:

Image Image


Next model was redesigned to use four common AA cells:
Spoiler
Show
Image Image

Electronic PCB marked "P&G AW61673/D" is densely packed with components, main one - Hitachi 6301 microcontroller. Other components are HEF40106BP inverting Schmitt trigger, HEF4069UB hex inverter, HEF4011B quad NAND gate, and a couple of BC308BP, C548A and ZVN2106A transistors.

Image Image


The purpose of reed switch present on PCB is not known to me. If actuated by passing the magnet nearby, the device emits some IR pulses. Maybe it's a some kind of "secret" service code, but I was unable to find related information in available Apricot technical documentation.

Switch on the bottom of device used to select between two modes: using a pair of infrared LEDs over the air, or separate LED if "wired" connection is used, to conserve the battery power. In that mode, trackball was connected to the computer via fiber optic cable, called "light pipe" in Apricot manuals, delivering IR signal directly to the PC receiver and insulating it from ambient pulses from "foreign" units, if more than one machine was used in the same room.

Image Image


Power consumption is 20mA in standby, activity in a "light pipe" mode with single LED takes 30mA, and wireless operation with two LEDs consumes 40mA. Data stream is transmitted synchronously as Hamming ECC encoded 4-byte packets at 3.85 Kbps rate.

Front panel is made of IR-transparent plastic. This is how it looks under normal visible light, and if illuminated by IR torch & pictured with infrared-capable camera:

Image Image


Device comes with included sheet steel bracket of complicated shape, pressing the ball against bottom bearings when attached. Probably it was intended to provide a resistance and suppress free spinning of the ball, if used in high vibration environment. None of Apricot documents I've found contain any mention of this thing.

Image Image


Pages related to the IR peripheral, extracted from ACT manuals and combined into single PDFs:
The Technical Reference Manual for the ACT Apricot F1 microcomputer
The Technical Reference Manual for the ACT Apricot Portable (Fp) microcomputer

Schematic diagrams of keyboard's IR transmitter and PC receiver circuits:

Image Image


Original package box:

Image Image


ACT and P&G advertising papers and reviews from 1984 - 1990:
Spoiler
Show

Image Image

Image Image

Image Image

Image Image

Image Image

Image Image

Unfortunately, this really fine device was created ahead of time: back then, nobody actually needed it for routine jobs, as existing applications were text-based and driven by keyboard - and a couple of years later, when GUI became popular, the market was already flooded by much more affordable alternatives. My sample is still working perfectly, and theoretically it's possible to use it in purely original form. Data format and encoding method are documented very well, so building IR receiver capable to decode the signals and convert them into USB HID packets is totally valid option. I've used USB Infrared Toy device with IrScrutinizer software to monitor the transmission:

Image Image


But IR was abandoned in PC peripherals for a reason: it's much less convenient than radio link. So, I'm planning to convert it to the 27MHz wireless USB device in the near future.