As an electronic hobbyist, having multimeters and having lots of them seems like a necessity. Well, no, not really, but I tend to get a bunch of them for the heck of it. I got this full sized Eico 555 (the 565 is similar, does not use 1% resistors), from a hamfest. Originally sold as a kit around 1960 makes it half-century old technology. The 555/565 are "31" range (I count ... 1xHVDC 1xHVAC 5xDCV 5xACV 5xdB 4xmA 1xA and 3xΩ = 25 ranges... where the 31 (on page 26) comes from I have no idea, perhaps the marketers were counting the nonexistent AC current ranges; the ammeter modes appears to be DC-only.) 20KΩ/VDC, 1KΩ/VAC, 50µA multimeters. A black, fairly heavy bakelite case encapsulates the meter.
|1000||V AC or DC (used for 5KV range by changing probe locations).|
Note that the "OUTPUT" selection can be used for dB for all voltage ranges
|250||V AC or DC|
|50||V AC or DC|
|10||V AC or DC|
|2.5||V AC or DC - I was wondering about 2.5VAC, but Ge diodes explains the meter markings|
|10m||A DC (Use this setting for 10A DC as well, but make sure you move the probes into the correct jacks)|
|R x 1||ohms (passes approx 111mA with 1 ohm resistor)|
Note that this is enough to forward bias single diode junctions (1.5V) and can destroy delicate semiconductors. It will supply enough current to light up "Joule Thief" LED power supplies.
|R x 100||ohms (passes approx 1.18 mA with 100 ohm resistor)|
|R x 10K||ohms (passes approx 64.3 µA)|
This meter is 4½" analog, and has just one semiconductor element in it - a dual diode half bridge for rectifying AC. It has batteries inside to deal with measuring resistance on its logarithmic (or is it 1/x, or is it something else?) scale. A UM1 "D" battery is used for the Rx1 and Rx100 scales. That D cell and 4 UM3 AA's are needed for the Rx10k scale. Its voltage scales are 20KΩ per volt impedance (DC) and has "change the probe location" 10A and 5000V scales. AC impedence is not as high, unfortunately.
The knob on the right tunes the scale to 0 ohms when the probes are connected together. Since analog resistance is an imprecise measurement, zeroing is needed each time. This differs from digital meters that have power for an internal current source, which let people read resistance as a number directly. If current sources were easy to make precise in the 70s, then there could have also been direct read ohmmeters; alas this didn't happen.
I noted that the weight of the meter movement does affect accuracy. I think this meter may have to be placed flat on the ground to be most accurate. Using it as pictured, upright, affects the reading.
The meter that it uses, since it is 20KΩ per volt, needs to be of 50µA which it is. This means that it takes 50µA to deflect the meter full scale from its terminals. Using the 100µA scale, half the current goes through the meter, and the other half through resistors in the body.
After the hiatus and a bunch of searching, I finally found the schematic for the meter. That strange battery holder wasn't really that strange at all, it was actually a plain old D (UM1) cell holder and is the main battery for the resistance modes. With the schematic, I could deduce how to wire it back in to the meter. With it hooked up properly I can see the meter deflect when I hook up a resistance to it for the first time. The resistance modes were still not totally working (really inaccurate) and the AC modes was not doing anything to the meter.
I subsequently found the diode halfbridge used for AC rectification was fried
(it was shorted) and the 11.5Ω resistor at the bottom of the resistance
measurement stack was around 30KΩ -- way off.
Replacing these two units (I replaced the diode halfbridge with two 1N34A
Germanium diodes, and the 11.5Ω 1% resistor with a 10Ω 5% and
a 1.5Ω 5% resistors in series with the assumption that I don't really
care about accuracy, the meter appears to be working once more.
Now measurements are very close (within 5% or so of my digital multimeters):
it now actually measures AC, and resistance is now deflecting more like what
the meter markings indicate.
I tried a few resistors from my junk pile and they all seem to either
be within the ballpark, or some even spot on the printed value.
Last thing is just that DC 10 ampere range is still measuring about 0A.
It does deflect the meter a little, just not enough -
perhaps the shunt resistor is not correct?
After the final bit of debug, I concluded the shunt is indeed wrong. I calculated the resistance of the shunt to be around 2.22 milliohms. This is an order of magnitude off from the 25 milliohms specced. I ended up simply ganging 8 0.2Ω 5% resistors in parallel to get 25 milliohms. Replacing the shunt with this blob of resistors, the ammeter works fine now. Remember that the probes need to be moved into the "±10A" jacks just underneath the meter face, and the select switch in the 10mA/AMP mode. Be careful not to forget to move the probes to the ampere jacks before measuring else it's one more way to need a new multimeter.
If you need the schematic, you can pick up a copy at Ken Kaplan's Electronic Kits website. A copy of it is here.
Oct 22, 2013 - New photograph taken by a DSLR instead of a camera phone, and changed text wording a bit.