Goetz Romahn's Homepage

Last updated 2005-1-23

The first HRPT receiver I constructed consists of two main components:
Downconverter and Receiver

Downconverter

For HRPT reception I use the downconverter developed by Radek Václavik. A kit may be purchased from Miroslav Gola. This downcoverter is easy to construct, easy to align and very sensitiv. Kits are not too expensive ( you could say cheap ). Frequency translation is from 1691 MHz to 137.5 MHz only.

In addition I also do own a downconverter published in UKW-Berichte ( VHF-Communications ) developed by Bernd Bartkowiak some 15 years ago and sold as a kit . This converter is a two-channel converter for input frequencies of 1691 MHz and 1694.5 MHz. Since this converter does not work as well for HRPT/HRI reception I use it only for Meteosat APT. May be some gain is missing in ther receiver chain since noise should not be a problem when using a good preamplifier.

Receiver

My first ( and still most versatile ) receiver is the HPR137, designed by Jaap van Rotten and Arne van Belle. Special parts ( eg. coils, SAW filter, tuner module etc. ) are available from Barend Hendriksen. The printed circuit board I got from Hans Beukinga (sorry no longer available from him). I made some modications to the power supply by including a DC/DC converter to generate the 33V tuning voltage needed and to the ZetF channel switch. This switch circuit I replaced meanwhile by a PIC microprocessor controlled switch which is in addition to switching receiver channels also capable to switch my Alblas C-HRPT/PDUS decoder to its proper operating mode accordingly.
Schematic if the PIC controlled channel switch ( click to enlarge).

You will find all the necessary details for construction and programming of the PIC with source codes included on my download page. I also included means for remote control via RS232 serial line ( line speed 19200bps ). An associated DOS utility program for testing or to select channels via a serial port is also available.
From a user point of view this controller is fully compatible with my PIC controller used with the RIG HRPT receiver ( see below ). So if you got your mouth watered by reading Bob Barnes article in RIG-Journal VOL. 71 to set up a fully batch controlled HRPT station by using David Taylors Wxtrack software, this will be ( and is in my system ) also possible with the "Kunstmaanen" receiver.

I put my HPR137 receiver and Alblas decoder into the same enclosure ( for an open view see below ). On the left side you see the bottom part of the open enclosure with receiver in a metal box an the tuner module mounted to its lower side. The channel switch, some status LEDs, field strength and tuning meters are monted to the front panel ( left side of the picture ). Part of the wiring needed some extra screening due to the neighbourhood of the digital decoder module when lid is closed. On the right side you see an inner view of the lid with decoder and DC/DC converter, to provide -12 Volt for the receiver, mounted. When looking at the front panel, you may be missing channel select switches and corresponding indicator LEDs from the decoder. I do not have and never missed them while using Alblas HRPT software. There are also no "mode switches" to switch Alblas decoder from HRPT to either CHRPT or PDUS mode and back again, since this is done automagically by my new PIC controlled channel switch ( see above).

This receiver is now in regular use at my home for receiving HRI transmissions from Meteosat 7. With my 135cm prime focus dish antenna and a Vidmar preamplifier I receive nearly speckle free pictures 24h hours a day ( the nonencrypted of course ). For an example see my images page.


HPR137 front panel



HPR137 inside view, receiver board in metal can on the left, Alblas decoder bord botton right

In summer 2002 I complemented my receiver equipment by constructing the RIG HRPT receiver developed by Sam Elsdon and available for members as a kit from RIG. Details of adventures during construction and about some modifications I made will be described below

Sam Elsdon's RIG HRPT receiver

before continuing to my notes I want to point out that I am not !! going to blame somebody, especially not Sam Elsdon. I sent him my notes and he commented carefully to all of them. Thank you Sam, thank you for this interesting design and thank you to the RIG-team for making this KIT available.

1. Kit arrived complete with all parts included. Perfect !!

2. While constructing the RX following the step by step method from Sam's manual, I had a problem: the PLL did not lock. After a long search that needed a spectrumanalyzer to identify signals, I detected a short circuit between pins 7 and 8 of the premounted synthesizer IC. I removed that carefully under a magnifier by using a very sharp knife ( actually a scalpel ). Now the PLL worked as expected -- occasionally ! The reason was a bad contact of pin 16 to the power supply. I seems, that this IC had been misaligned during fitting to PCB.

3. Further construction proceeded smoothly, thanks to the manual.

4. The programming error of Fengyun frequencies in the first release of PIC software Sam already detected and corrected. It is also corrected in my modified software. See below.

5. One problem still remained: there is a drift in the demodulator PLL that makes the RX rather difficult to use. See Sam's website for any countermeasures.

6. To handle the drift problem I did the following ( see excerpt of schematic ) :

- I added an adjustable voltage via 15 kOhm to pin 6 of U13B. So I can finetune the VCO frequency from the frontpanel.
- to gain some feedback I connected a zero indicator instrument to the I and I_ pins ( +- 100uA indicator with an internal resistance of 1750 Ohm ). Connecting this instrument did no harm to the signal quality.
- additionally I developed a lock indicator by connecting an op amp to the Q and Q_ outputs of the demodulator similar to U13B beeing connected to I and I_ . The signal level for lock-detect is determined by adding an adjustable voltage to the inverting input of that op amp. The output of the op amp is connected to an LED. This lock detector has proven to be very sensitive -- signals are detected when the S-meter does not even move a little -- and reliable. So one may omit the zero indicator mentioned above ( I did not, I like it very much )

7. R15 from the schematic is 1k, a rather large value. I added another 1k in parallel resulting in 500 Ohm. This gives higher gain and better large signal performance.

8. I migrated the controller software from PIC16c54 to PIC16F84. For this to work at least R7 should be >5k and C17 abt. 100pF.

9. I added Meteosat frequencies of 1691 MHz and 1694.5 MHz . This makes it possible to use Meteosat as a beacon for testing the whole system and (!!!) for fine tuning the demodulator VCO frequency with a stable reference frequency. I would strongly recommend to put in Meteosat frequencies into the next release instead of the unused 1704.5 and one 1707 channels.
BTW I received HRI pictures from Meteosat with this RX and an 135cm prime focus dish antenna with only few speckles and bad lines. Sourcecode for this minimal modification with migration to 16f84 and Meteosat 7 frequencies is available here. Do not forget the necessary modifications of point 8.

10. The PIC software has been modified additionally to have serial input for remote control instead of the time-step interface and control of the Alblas decoder settings for the various data formats of the different satellites. This modification results in a single switch operated receiver for HRPT, C-HRPT and HRI ( less VCO fine tuning ). Source code for this with an associated DOS utility program to select channels via a serial port is also available. Some modifications to the hardware need to be done for this software to work also. Since I am not going to use time-step remote control I removed the resistor network R12 and cut pin 20 of U5 . Connector J3 now has spare pins which I use to connect to the serial port and to the Alblas decoder.

11. The IF chain with L3 and L4 needs slightly different settings for HRPT and C-HRPT, so you have to find some compromise.
You will find a complete package of my amendments to the RIG HRPT receiver on my download page.

What have I achieved so far ?

I tried to assemble a very small HRPT receiver station ( the second smallest of the world ?) also good for mobile use with +12V battery power. To achieve this I squeezed the RIG HRPT RX and my Alblas decoder board into a single small aluminium box ( see pictures below, meter scale is centimeters not inches ). On the front panel I have meters for signal strength and fine ( zero) tuning as well as indicator LEDs to show demodulator lock, bit errors, frame-sync and LPT-port status, channel select switch and frequency fine tuning. At the rear panel are connectors for PC-LPT port, serial port for remote control, C-HRPT eye signal output, RF-input from antenna preamplifier and +12V DC.

The picture below is a top view to the open box with decoder board removed. I used a smaller heatsink for the power regulators than the original. The perfo-board contains the lock detector. It is mounted piggiback to the PCB riding on some jumper posts. Cable has been given some added legth, so the unit can be disassembled for testing purposes. Since my "bad lines problem" with my Alblas decoders have been settled recently, all nuts and bolts have been securely fastened and this receiver is in regular use at my home.

Back