Heathkit IM-102
A cold-cathode digital multimeter that put lab-grade measurement on the hobbyist's bench - if the hobbyist had ten hours and a steady soldering hand.
Overview
Flip the range switch on an IM-102 and the reading doesn't waver. No parallax, no squinting at a needle. Cold-cathode readout tubes light the digits through a smoky plastic window, and the display refreshes roughly five times a second without blinking. A "memory" circuit holds the number steady so it reads the same way for everyone. That was the pitch in 1972, and it was a real one. This was the first digital multimeter Heath offered in kit form, and Electronics Illustrated noted at the time that until Heath made a DMM available as a kit, the device's "sole domain was the lab." Audio (November 1973, p.62) described the conversion scheme as a voltage fed to a linear IC whose output charges a capacitor discharged at a constant rate, with the discharge time measured by a counter.
The kit was not cheap. Heath's own 1976 catalog lists the IM-102 at $249.95 shipping weight 9 lbs, while the earlier Electronics Illustrated coverage pegs the kit at $229.95. That magazine said it seemed expensive until you considered that the finished instrument cost less than half what an off-the-shelf digital multimeter of equivalent performance commanded in 1972.
DC volts run five ranges from 200 mV to 1000 V. Input impedance is around 100 megohms on the 200 mV range and roughly 1000 megohms on the 2 V range. That makes it useful for high-impedance circuit work where an ordinary VOM would load the node under test. Higher DC ranges drop to 10 megohms. Accuracy is specified at ±0.1% ±1 digit when calibrated against lab standards, or ±2% ±1 digit with the furnished DC calibrator. Heath printed that two-tier spec plainly: the IM-102's accuracy depends on what you calibrate it against.
Resistance measurement covers six ranges from 200 ohms full-scale, where it resolves down to 0.1 ohm, up to 20 megohms. AC volts run five ranges from 20 mV to 1000 V, average-responding and rms-calibrated, with 1 megohm input impedance at 150 pF. Current measurement covers 200 µA to 2 A on both AC and DC, with a nominal 0.2 V voltage drop.
The calibrator bundled in the box is what made that accuracy achievable. Heath shipped a wired, factory-adjusted unit built around a mercury cell and precision voltage divider, set to exactly 200 mV DC. The magazine's testers checked their supplied calibrator against a standard cell and found it within 0.1 percent of 200 mV. After calibrating the meter with it, they found the resulting accuracy met or exceeded Heath's own published figures.
The conversion scheme is a patented dual-slope integrator that, per Heath's literature, does not depend on a stable clock frequency for accuracy. That removes one of the more finicky alignment steps for a kit builder. Overrange is indicated automatically beyond a count of 1999, with a front-panel light. Polarity is displayed automatically as "+" or "-", so there's no need to swap probes. The whole thing runs on 120 or 240 VAC at 8 W nominal, with a dual-primary transformer and an isolated floating ground that Heath rated for operation up to 500 volts above power-line ground during line operation. The cabinet is aluminum with a die-cast zinc front panel, and the detachable three-wire line cord means no batteries to leak.
Building one separates the IM-102 from its analog sibling, the IM-105. The IM-105 assembles in a couple of hours and is meant for beginners. The IM-102 runs to roughly ten hours and demands care. The magazine's builders found several unfilled holes in the board with no explanation, though the finished kit worked fine. A working IM-102 on the used market suggests a careful builder, and a non-working one often has a solder bridge or a cold joint hiding somewhere.
Calibration is the other maintenance reality. The spec sheet gives two full accuracy tables, one for lab-standard calibration and one for the furnished 200 mV calibrator, and the difference is substantial on some ranges. With the DC calibrator, resistance accuracy is ±0.5% ±1 digit on the 200, 2k, and 20k ranges and ±1.5% ±1 digit on 2M and 20M. With lab calibration those tighten to ±0.25% ±1 digit on the 2k, 20k, and 200k ranges. DC current accuracy with the calibrator is ±0.5% ±1 digit for 2 A but ±3% ±1 digit on all other ranges. Under lab calibration that tightens to ±0.25% on the 2 mA and 20 mA ranges. A unit calibrated with a tired mercury cell won't deliver lab numbers.
Normal-mode rejection is specified at 35 dB minimum at 60 Hz and common-mode rejection at 80 dB minimum with 1k unbalance at 60 Hz. That is adequate for bench work, not a substitute for a guarded lab meter. The display maxes at 1999 counts, the operating temperature window is 10°C to 40°C, and the case measures 3" H x 7" W x 7.9" D.
eBay Listings
As an eBay Partner, we earn from qualifying purchases. This helps support our independent vintage technology research.