[hpsdr] Problems Calibrating Pennylane and with LPU, Could Use Some Help

[Bob Mahrenholz]

Hi:

I've been trying to calibrate my new Pennylane card HPSDR per the 
instructions on Page 12 of the Penelope Manual.
(http://openhpsdr.ogr/wiki/images/4/4c/Penelope_manual_v_1-2.pdf)

I have just installed the Pennylane card in my HPSDR Pandora which 
has Magister and Mercury all powered from the LPU.
I installed the 17-Nov-2011 version of PowerSDR1.16 from W5WC (I 
prefer its look to the "skin" of PowerSDR1.19) and have
no problems in receiving with Mercury. Firmware versions are 18 for 
Magister, 29 for Mercury and 13 for Pennylane. I am using
Winows XP SP3 on an Asus based PC.

I have connected a 50-ohm dummy load to the Penelope output BNC.  I 
have my scope connected across the load resistor so
I can measure the output voltage and its frequency and look at its 
waveform. I do not have a microphone connected as yet.
The "Drive" level is set at 100. Everything seems to perform as 
expected except for the output power adjustment.

A problem occurs when I try to run the calibration routine in the "PA 
Settings" Tab as recommended on Page 12 of the
referenced manual. The initial "Gain By Band" settings are all set to 
49.0 dB. When I click on "Calibrate", Pennylane
starts stepping thru increasing amplitude signal values at 160 meters 
until at the 4th step the signal begins to overdrive
the output of Pennylane and the signal on the scope shows clipping. 
After a couple of more steps, the routine stops with
an "Invalid Gain Found" message. The clipping level is just under 20 
volts peak-to-peak (0.5 watts). The amplitude of the
output first step of the calibration measured 10v p-p on the scope, 
the second 15v p-p, the third 16v p-p and the 4th and
following steps are heavily clipped at 18v p-p. The TX Meter "Fwd 
Pwr" read successively 251, 408, 624, 628......mW.

After trying various settings,  I found that if I crank the "Drive" 
down to 4, the calibration runs without bombing. In this case
the amplitude of the output remains constant as the frequency is 
stepped to each band in turn. However, after the calibration,
the "Gain By Band [dB]" values are not changed from their initial 
settings as I expected they would be.

Since Pennylane was checked out at IQLabs, I assume I have not set 
something up right but I can't figure out what it is.
Or does Pennylane not work the way Penelope does?

Can someone explain just what is being calibrating and how it is 
done. Apparently Pennylane is feeding different input signal
levels into its input and is reading the output so as to set the 
system gain on each band to get the same output at each rf
frequency. Also, what is the "Advanced Calibration Technique" that is 
selected by checking the box in the "PA Settings" Tab?

I have found that I can do the calibration manually. For each band, I 
set the "Gain By Band [dB]" to 49.0 and slowly crank it down
until the "Fwd Pwr" as read on the TX meter reads 500mW. This is with 
the "Drive" level set back to 100. This appears to work
properly and gives me a consistent 0.5 watt output on each band.

However, if I hold the output at 500mW for too long, I hear a "poof" 
from Pandora and everything goes dead!  All the red lights
go out and the program stops. If I turn my power supply off and back 
on again, things seem to recover. I suspect maybe a fuse
blowing because of excessive current drain but the total Pandora 
current drain including the fan only measures 1.670 amps. LPU
has 2 amp fuses in both its 12 and 5-volt outputs.

I have no experience with the type resettable fuse used in LPU. Do 
they make a noise when they blow? Are they fast acting or
slow acting? They are 2-amp fuses (unless I got them mixed up) with 
only 1.670 amps thru them. What sort of accuracy may be
expected for their blowing current? Maybe I should measure the supply 
current on my scope and see if there are any high
transient currents. Anyone's experience on these matters would be most helpful.

Thanks for any information or any ideas.

73,
Bob Mahrenholz
K4QQK











-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20111128/3febf1a9/attachment-0003.htm>