[HBR] HBR-16 Front End Experiment

[LeeCraner at aol.com]

Recently, I did some experimenting with the front end of my HBR-16, and  
came up with some interesting results.  I thought the reflector group might  
be interested.
 
Summary Results:
After replacing the typical three terminal (antenna/antenna/ground) screw  
type antenna connection on my HBR-16 with a coax connector, and then feeding 
the  primary of the HBR-16's input coil, L1 directly with RG58 coax, the 
band to band  variation in the receiver's input impedance was substantially 
reduced.  The  resulting input impedance, by band, before and after the 
modification was:
 

80   94 ohms (was 94 ohms)
40   75 ohms (was 75 ohms)
20   84 ohms (was 33 ohms)
15   68 ohms (was 27 ohms)
10   (see text)
 

My receiver: 
My HBR-16 front end was constructed per Ted Crosby's original  design.  
That is, the input coil L1 is mounted close to the front panel and  the antenna 
terminal is a ceramic three screw terminal on the back of the  chassis.  
The primary of L1 is connected to two of the antenna  terminal screws through 
two coax cables giving the HBR-16 a balanced  input.  The coils were wound 
according to Ted's article.
 
In typical modern fashion, I use unbalanced 50 ohm coax from the  receiver 
to the antenna.  So the center of the coax is connected to one of  the 
antenna terminal screws, the shield is connected to the other antenna  terminal 
screws and a jumper connects this screw to ground. 
 
Object: 
It has been suggested that a 300-75 ohm TV balun improves many older  
receivers that have a Hi-Z balanced antenna connection, such as the 75A Collins  
receiver, when connected to a coax fed antenna.  The original object was to  
calculate the input impedance of my HBR-16 to determine whether such  a 
balun, connecting my unbalanced 50 ohm antenna to the balanced HBR-16 L1  
primary would be worthwhile.  In other words, if the HBR-16's had  a Hi-Z antenna 
input impedance at the antenna terminals (200 ohms or more), a  4:1 balun, 
matching the unbalanced 50 ohm antenna to the balanced L1 primary,  might be 
worthwhile.  
 
Equipment Used:
URM-25D RF signal generator (the URM-25D has a 50 ohm unbalanced  output)
CN-224 50 ohm 20 dB (10:1 voltage) pi attenuator
MX-1487 50 ohm termination (aka "impedance adapter")
Tektronix 465 two channel ooscilloscope
 
Theory:
By comparing the RF voltages at the input (Vi) and output (Vo) of an  
attenuator between an RF generator and an unknown impedance, the unknown  
impedance, Z, can be calculated using:
 
Z=(15,225Vo)/(61.1Vi-308.6Vo)
 
Where 61.1 ohms is the input and output shunt resistors of the attenuator,  
247.5 ohms is the attenuator's series resistor, 15,225 is the product or 
61.1 x  247.5 and 308.6 is the sum of 61.1 + 247.5.
 
Test setup:
In order to measure the input and output RF voltages at the attenuator,  
coax tee fittings were connected at each end of the attenuator and the output 
of  the two tee fittings were connected to the two channels of the  scope.  
 
The output of the signal generator was then first connected  through the 20 
dB attenuator to the 50 ohm termination and  measurements taken in the 
middle of the five primary ham bands.  At 80, 40,  20 and 15 meters the output 
voltage at the attenuator was exactly 1/10  (-20dB) of the input voltage, 
thus validating that the setup was correctly  seeing the 50 ohm termination 
impedance on these bands.  However, at 10  meters, the output voltage was about 
1/20 the input voltage, thus  invalidating any results on 10 meters 
(probably due to length of the connecting  cables, etc.).  Since I don't use my 
HBR-16 on 10, I confined any further  testing to 80-15 meters.
 
The 50 ohm termination was removed and the attenuator hooked up to the  
HBR-16's antenna input terminals.  Measurements in the middle of the 80-15  
bands were again taken and the HBR-16's impedance at each band calculated from  
the input and output RF voltages. 
 
The results, by band, were:
 
80 94 ohms
40 75 ohms
20 33 ohms
15 27 ohms
 
With these Lo-Z results, it appeared that a 4:1 balun (the original reason  
for the experiment) wouldn't help match a 50 ohm antenna to the  HBR-16 and 
that a direct 50 ohm connection at the L1 primary might be a  better route. 
 So I removed the three terminal antenna block and the two  coax cables 
connecting it to L1's primary.  In its place,  I mounted a BNC jack to the 
receiver's rear chassis and connected it  directly to one side of the L1 primary 
through a length of RG58 coax.  The  other side of L1's primary was 
grounded at the L1 socket (in Ted's original  configuration this would have been 
grounded at the rear chassis when using an  unbalanced antenna).  Remeasuring 
the impedances at each band at the  newly installed BNC connector resulted 
in:
 
80 94 ohms
40 75 ohms
20 84 ohms
15 68 ohms
 
Conclusions: 
It would appear that the original balanced configuration with two coax  
cables between L1 and the antenna connector significantly affected the  
impedances on 20 and 15, and that bringing the 50 ohm unbalance line all the way  
to L1 had some benefit.  Though no results were calculated at 10  meters, it 
probably benefited from the modification as well. 
 
A question remains whether the modification resulted in any practical  
improvement in performance.  Unfortunately, I didn't measure the  sensitivity of 
the receiver before the modification.  I did a quick  sensitivity 
measurement on 40 meters after the modification (40 meters being the  coil set that 
happended to be installed at the time) and got a 0.32 uv 10dB  S+N/N figure.  
Not bad for a 50 year old homebrew receiver design.
 
If anyone would like a copy of the Excel spreadsheet I used to make the  
impedance calculations, drop me an email.
 
73
Lee WB6SSW