[time-nuts] Vintage Frequency Measurement

Mon Feb 13 05:24:59 UTC 2017

On 02/12/2017 01:08 AM, Scott Stobbe wrote:
> I was inspired recently coming across a Lampkin 105 frequency meter, as to
> how  frequency measurement was done before counters.
>
> Certainly zero-beating a dial calibrated oscillator, would be one approach.
>
> Is there a standout methodology or instrument predating counters?
>

Hi Scott.  That Lampkin 105 is a sophisticated design.  I did 
some research that you might be interested:

For the 2015 November ARRL Frequency Measuring Test, I fired up 
my old Lampkin frequency meter.  For their 100th Anniversary, 
QST was encouraging the use of “vintage” equipment for the FMT, 
and the Lampkin was designed in the 1930s.

I (AA8K) did surprisingly well, coming within 322 Hertz on 40 
meters, 202 Hertz on 80 meters, and 18 Hertz on 160 meters.

The Lampkin 105-B was designed by Guy Forest Lampkin BSEE, who 
got his first ham license in 1924.  In 1933 he was selling the 
model 102, that was checked with the Federal Radio Commission 
and commercial laboratories to be within 3 to 15 cycles at 1,712 
kc.  He was also selling a “foundation unit” of the Precision 
Micrometer, Band Spread condenser, Special Isolantite coil form, 
Temperature compensator, Adjustable pad condenser, and complete 
circuit details for $14.50.  Lampkin Laboratories moved from 146 
West McMillian Street, Cincinnati, Ohio to 8400 Ninth Avenue 
N.W., Bradenton, Florida 33506 in 1935.  It was incorporated in 
1942.  Precise Power Corporation had acquired Lampkin 
Laboratories in 1971/Oct.  At that time Lampkin Labs had 17 
employees and wasn't advertising their 107B Digital Frequency 
Meter because they were selling as fast as they could make them. 
  The last known address was 12297 US Highway 41 North, 
Palmetto, Florida 34221.  Voluntary Dissolution 2007/April/27.

The 105-B is a fascinating design, able to measure frequency to 
0.0025%.  Signals can be measured from 100 KHz to 175 MHz.   It 
works similarly to the later World War II BC-221 frequency 
meter.  It can receive, or transmit the internal oscillator 
2330-2670 KHz.  A diode generates harmonics that can beat 
against the signal to be measured.

The variable condenser has a precision-machined tubular stator 
and a tapered, conical rotor.  They are made from steel and 
brass and copper plated.  The parts are proportioned such that, 
due to the differential thermal expansion, the temperature 
coefficient of capacity is a few parts per million per degree 
Celsius at all positions of the rotor.  The rotor is moved in 
and out of the stator on a micrometer screw.  The large dial and 
turns counter give a dial band spread of 8,000 divisions over 42 
feet!  The inductor is wound on a six-ribbed form of 
polystyrene.  Since the thermal expansion of polystyrene is 
greater than copper, as coil temperature increases, the turns 
are pulled from circular to hexagonal, and the average diameter 
of the coil decreases.

Thermal design is utmost in the Lampkin MFM.  In addition to the 
L/C circuit, the vacuum tubes and circuitry are mounted on the 
rear, with the chassis cut-away to keep it from heating the 
front where the L/C and calibration crystal are mounted.  Wires 
connecting the L/C and crystal are very small diameter to reduce 
the thermal path.  Even the power transformer is bolted to the 
outside of the cabinet.  The 7.5 MHz calibration crystal (no 
oven) is held against the front panel.  There is a glass 
thermometer mounted to the front panel.  It is custom-marked 
with a correction factor for that specific unit.  I left the 
Lampkin turned on from October, but the temperature soon 
stabilized.  The metal 6J7 tube has the Bakelite cap removed, to 
eliminate changes due to moisture absorption in the Bakelite.

Striking features of the Lampkin are:  The very smooth tuning 
with almost zero backlash.  Turning the dial clockwise lowers 
the frequency, but that is because the micrometer screw is a 
right-hand thread and moves the rotor into the stator, 
increasing capacitance and lowering frequency.  Increasing 
frequency moves the micrometer post outwards through the center 
of the dial, like Pinocchio's nose growing.  The outboard 
transformer looks clunky at first, until you realize why he did 
it.

Modifications.  I replaced the 1 ampere line fuse with a 0.5 
ampere for additional protection.  (I have been wary of 
unattended equipment ever since an un-fused Radio Manufacturing 
Engineers receiver monitoring RTTY autostart almost burned our 
house down in the 1960s.)  Measuring frequency requires table 
look-ups and correction math.  Since I did not have the original 
manual with calibration data, and the end-stop was nowhere near 
000 on the counter dial, I decided to calibrate it myself.  At 
the same time, I added ferrite inside the coil to drop the 
oscillator frequency down into the 160 meter band.  Using my 
OpenHPSDR, I calibrated by noting the Lampkin dial reading for 
every 100 Hertz on the HPSDR.  During the FMT, I used a Collins 
75A-4 receiver in AM mode and adjusted the Lampkin 105-B to 
zero-beat the signal.  Using the Lampkin dial reading and 
looking it up in the table, I could interpolate for the 
frequency between the two calibration values.

During the test, I noticed that the Lampkin was varying because 
the line voltage was changing.  The next time I will build the 
recommended line voltage regulator using two 0C3 (VR-105) tubes 
and a 60 Watt ballast lamp.

I uploaded my mods, photos of the inside, and a scan of the 
manual to mods.dk.  I tried to ftp and email the manual to 
BAMA/eDebris without success.

See attachments
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Guy Forest Lampkin born 1905-Apr-02 Wolcott Indiana died 1972-May Florida.png
Type: image/png
Size: 277639 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20170213/d89ae3a9/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Lampkin ad 1940 compressed.jpg
Type: image/jpeg
Size: 57702 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20170213/d89ae3a9/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IMG_4582 FMT.JPG
Type: image/jpeg
Size: 283716 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20170213/d89ae3a9/attachment.jpe>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Lampkin 101 102 1935-September.jpg
Type: image/jpeg
Size: 50734 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20170213/d89ae3a9/attachment-0001.jpg>