The Wireless Set No. 19 Homepage
Observations - Frequency Meter - BC-221-AK By D.G. Lawrence, VA3ORP


1.         This report is based on observations made while using "Frequency Meter - BC-221-AK" serial number "1470: C.P.R. 34507-PHILA.-43".  This instrument was purchased at an amateur radio flea market for $25.00 (Canadian) in the late 1990s.  It is unmodified except for a removable, 115 vac, regulated power supply which resides in the battery compartment.


2.         The BC-221 is a three tube, heterodyne type frequency meter with an integral crystal calibrator.  It can be used to set the frequency of both transmitters and receivers over a frequency range of 125 KHz to 20 MHz.  The instrument has two ranges - "low" and "high".  On the "low" range it operates between 125 and 250 KHz on fundamentals while the "high" range covers 2 to 4 MHz.  All other frequencies are determined by using higher order harmonics.  The manual for this instrument indicates that the maximum error will occur when measuring a 4 MHz signal at minus 30 degrees C where an error of 1.355 KHz (0.034%) could occur.  In practice, only half of this error is likely to be observed.  Under more favourable conditions, much less error can be anticipated.



3.         The purpose of this evaluation was to answer two questions:

            a.         Is the BC-221 sufficiently accurate to meet modern needs for frequency setting in the amateur service? (i.e. accuracy)

            b.         Can the BC-221 be used to easily set the frequency of a Wireless Set No. 19 during on-air operations? (i.e. stability and handiness)



4.         In order to qualify for the Wireless Set Number 19 Group's Vintage Operator's Award endorsement, it is necessary to use an item of equipment on five occasions.  This criterion has been met by using the BC-221 both operationally and for extended bench tests.  Observations from those tests follow.


5.         25/26 June 2002 - Operation "CHIRP FEST I".  The BC-221 was used as the frequency standard for the 19 Set Net.  This was a field operation with power provided by a gasoline generator.  The main observation was that careful and frequent setting of the "Corrector" control was required in order to keep the instrument in a solid zero-beat situation.  It was apparent that the BC-221 was drifting; however whether it was the VFO portion or the reference crystal oscillator was not determined.  The amount of this drift was minor compared to the normal drifting of a W.S. No. 19.


6.         1 July 2002 - Bench measurements were made to determine the accuracy of the BC-221.  Repeated checks against a local AM broadcast station showed an error of +53 Hz at 960 KHz (Low band).  Repeated checks at 7.023 MHz showed an average error of +98 Hz (range 60 to 150 Hz).  A single measurement made quickly could have an error as great as 250 Hz at 7 MHz.  This could be reduced to 100 Hz by careful zero-beating and by rocking the “Mode” switch back and forth to get a consistent result.  To set the frequency closer than 100 Hz would likely require minor amendments to the calibration book.


7.         2 July 2002 - Further bench tests were done to examine the drift characteristics of the BC-221.  All tests were done at 7 MHz.  The VFO exhibited an unusual drift pattern.  For the first 10 minutes it was drifting at a rate of -210 Hz per hour.  From 10 to 20 minutes after power up, the VFO remained stable.  For the next hour the instrument exhibited a fairly steady drift of +75 Hz per hour.  From 60 to 90 minutes after power up, the reference crystal oscillator exhibited no drift but there seemed to be some instability in the order of +/- 25 Hz.  Accuracy of the crystal oscillator was checked at the 7,333.3 KHz calibration point.  The output was 90 Hz low with instability of approximately +/- 20 Hz.  A test was done to determine the resetting accuracy at 7 MHz.  Over a 3 1/2 minute period 10 measurements were taken with errors ranging from -123 Hz to +157 Hz.


8.         The conclusions from the tests up to 2 July 2002 are as follows.  A single, quick measurement could be in error by up to 250 Hz (at 7 MHz).  Careful adjustment and repeating the measurement three times could reduce the error to 150 Hz.  Half of this error is from the reference crystal oscillator and calibration book errors while the other half is from inaccuracy of dial setting, zero-beating and both "Corrector" control coarseness and "Mode" switch instability.


9.         22 July 2003 - Further drift tests were conducted.  On the afternoon of 21 July 2003, the instrument had been warmed up for about one hour and then carefully set to 3,705.000 KHz.  After cooling down overnight, the set was turned on and the calibration controls were left untouched.  Output frequency was then measured over a two-hour period.  The initial frequency was +500 Hz (i.e. 3,705.500 KHz).  After ten minutes the error was +250 Hz.  At 50 minutes, the error was +60 Hz.  From this point until the end of the test, there was a steady drift at a rate of 210 Hz per hour, ending the test at 3,705.270 KHz.  The conclusion from this test is that the set needs at least a ten (10)-minute warm up period and 30 minutes would be preferable.  If the instrument was carefully set after 30 minutes, the error for the next hour would be only +/-50 Hz due to drifting.  The reason for the difference between the drift rate during this test (210 Hz/hr) and that noted on 2 July 2002 (75 Hz/hr) was not determined but may simply be due to the second test being longer.  


10.       22 July 2003 - The BC-221 was used to set the frequency of a W.S. No. 19 Mk III during a QSO with VA3GST.  This was done by connecting an insulated wire from the BC-221 Antenna terminal to run in close proximity to the back of the 19 Set's variometer.  Once the BC-221 had been set, it was not adjusted for the duration of the QSO (30 minutes).  Normally the BC-221 was left in the "Crystal" position.  When it was desired to check the frequency of the 19 Set, the BC-221 was switched to "Operate", the 19 Set was set to "R/T" (receive) and the "Flick Adjust" was used to zero-beat.  Then the BC-221 was returned to the "Crystal" position, the 19 Set was returned to CW and the QSO continued.  VA3GST observed that the 19 Set remained on frequency during the entire QSO.  This was verified by his Kenwood TS-440 using the "Notch Filter" method which normally produces results within +/-30 Hz.  Note also that this QSO was done under "Blackout" conditions - i.e. the 19 Set operator was blindfolded throughout the entire QSO.


11.       23 July 2003 - The BC-221 was used during the 19 Set Slow Speed Net (19SSSN) for frequency setting.  This was also a "Blackout" operation.  The BC-221 proved to be fully capable of being used in this role.  At one point, the operator who was using the BC-221/19 Set combination identified a difference between his frequency and that of the Net Control Station.  An enquiry to the Net Control Station confirmed that NCS had drifted about 500 Hz from the original frequency.  Over the one (1) hour duration of this QSO, the total drift of the BC-221/19 Set combination was 300 Hz without any recalibration of the BC-221 having been done.


12.       26 July 2003 - The BC-221 was observed over a two and one-half (2 1/2) hour period to determine its accuracy.  Every 15 minutes the instrument was calibrated against its internal crystal and then the output frequency was checked using a Racal 6778C receiver with an output to a Tuning Indicator which had a CRT display with a Lissajous figure.  This configuration had been checked against the CHU time signal and found to be accurate within 1 Hz.  The test was carried out at 3,700.000 KHz.  After a 30-minute warm up period, the output frequency was 3,6999.965 KHz.  For the next two hours, the output frequency varied only +5 to -15 Hz from this frequency.  The conclusion from this test is that with careful zero-beating, the stability of the BC-221 is excellent having a maximum error of 60 Hz.  This accuracy could be improved if an adjustment were made to the internal crystal oscillator trimmer capacitor.



13.       The above tests support the following conclusions:

a.  As issued, the BC-221 is easily capable of 250 Hz accuracy on 80 & 40 M

b.  A warm up period of 30 minutes should be allowed

c.  Careful zero-beating is required to ensure accuracy

d.  Some error is caused by the instability in the Mode switch

e.  Some error results from inaccuracies in the calibration book

f.   The "Corrector" control is a bit coarse and requires careful adjustment

g.  When switching modes, about 10 seconds is required for the freq to stabilize

h. The BC-221 can be used to easily and accurately set the frequency of a 19 Set

i.   The BC-221 can be used to monitor drift during WS 19 on-air operations



D.G. Lawrence

VA3ORP – 26 July 2003  


Extract of email from Vic Politi (W1NU) on Friday 10 November 2006: 

"Dave asked John if he was using a sidetone device while transmitting with
his 19 set and John admitted, yes I do use one of those Wimp devices but
then on his next transmission he shut it off and asked Dave, how he copied
it. Both Dave and I copied it fine and did not notice any change in John's
keying. Dave then asked me what I used and I answered that I use a BC-221
with a small speaker to copy the 19 set signal. I warm the BC-221 and the
19 set up about three hours before net time so by the time the net starts
they are still pretty well stabilized. Amazingly the BC-221's that I have (I have five of them)

are all extremely accurate and they fit in nicely with my WW II setups.

Dave then asked me if we used them in the field during combat in WW II and
I answered in the affirmative. We had the field units in wooden cases
painted olive drab and they were powered by batteries. They even could be
used as an emergency transmitter. In North Africa, just before the Sicilian
invasion we had one last big practice landing and when we got ashore our
SCR-284 field TX was on the fritz. The receiver worked OK so we rigged up a
Windom antenna, using field telephone wire, about 130 feet long with an
off centre single wire feeder. Set the BC-221 up to the net frequency and
reported in to the net control station, who was about 1/4 mile from us and
got a QSA5/QRK5 back from them. To report in I just took the end of the
feed line and keyed the oscillating 221 by simply touching the antenna post
on the 221 and voila we were in the net. We were at Battalion HQ and the
NCS was at Regiment HQ with a BC-199/BC312 setup so they had no problem
copying us. We then used the j-47 key from the 284 by inserting it in the
antenna feed line. To receive we had to shut the 221 off as it would block
the 284 receiver and as the 221 has no B plus switch we simply switched the
221 switch on the front panel from Osc. to Xtal. only.

A few years ago, Joe, GDZ and Bob, KF1J, and I, had a CW net on 80 meters every
evening. They were using rice boxes and I decided to use the BC-221 Trick.
My G5RV apex is up at 55 feet so I figured it should work OK.  At first
they did not hear me as they were using sharp filters and with 100 watt rigs
had their RX gain turned down.  Finally Joe heard my pipsqueak signal and
alerted KF1J when they both turned up their gain I got 5-7-9 reports. They both live a
quarter of a mile from me, one to the east and one to the west.  My best DX
using it on transmit and receive was in the late 1950s with W1WX, one mile
from my QTH using his BC474 mobile at Fairfield Beach. 


Dave, ORP, makes these queries and I turn them into war stories, hi"


Photos - Frequency Meter - BC-221-AK

Return to Tech Notes

Return to The Wireless Set No. 19 Home Page

Site Copyright © by Bob Cooke VE3BDB, Orillia, Ontario, Canada.
All rights reserved.
Contents may be downloaded or displayed elsewhere ONLY if credit is given to the author and to "The Wireless Set No. 19" website.
Thank you for respecting these conditions.