Just as a reference - I was using RS15 (and its ancessors) during the "good old times" - in the old days just 25W on a 1/4 groundplane was more than sufficient to work the RS birds on the uplink.
But this might be different now of course.
Keep us informed about the progress.
73, Oscar
Yesterday I fired up mine (BLD6G22L-150BN2 variant) for the first time without smoke, hi hi.
I just tried it first in unmodified way. (only replaced the Input and output connectors, nothing else so far)
All Bias currents looked just fine as described here and (so far) no problems with unstable internal bias circuits.
But this could still happen, like others have experienced...
The amp produced nice strong signals, hi hi (I drove it up to approx. 8A at 28V, but cannot check output power right now).
However I noticed that audio was not sounding very clean as with my other amps (SG Labs or PA3AXA), a bit rough, specially when underdriving the amp for lower power output. Could this be related to the Doherty mode or to the Bias being optimised for contiunous data modes?
Regarding snowflaking I need to upgrade my measurement devices first, so I am not really ready for that yet (or at least have it calibrated against a good reference).
I wonder if the modification of the Doherty circuit is worth it? And perhaps also externally biasing it?
DB8TF In an older post you decribed your biasing circuit, but would be great if you could draw it on a piece of paper and post it here for more clarity?
Someone also mentioned (I think it was Sandor DM4DS on some other channel) that the 1st stage amp presents a total system limitation on 2.4GHz and is partially responsible for the -10dB amplification many people see on 2.4 vs. the original design frequency.
Can someone confirm this is the case?
So one of my next first things would be to remove it and do a direct connection to the second stage (similar as in YO5ER photo and others). It seems to me there are still some hams playing with this amp type so all hints published here could help others. Thanks for all the great posts so far.
73, Oscar
G0KLA is a very accessible guy (I've been in touch in the past regarding PACSAT logfile formats). I am prety sure he could perhaps supply a special version for QO-100 experimenting...
You could try to contact him.
73, Oscar
Based on the pictures I would agree that the skew angle is not set correctly, but above all I would not really trust that adjustable LNB arm either. It seems that not only the angle is variable but also the height? (since it uses a longitudinal hole) ?
Definitievely worth tweaking after you made sure that the azimuth is 100% corrently set to maximum signal.
When using the SDR in wideband mode (e.g. displaying around 1 - 1.5 MHz bandwith of spectrum) you should clearly see with your dish size at least 1-3dB noise level incease in the middle of the passband as compared to the noise levels seen on the spectrum display outside of the passband)
73, Oscar
PY1SAN You looked to the wrong datasheet section...there are many versions of this type 
This letter code is the CMOS version and it has output 90% VDD for logic 1 and 10% VDD for logic 0....that means close to 3V output, hence Michael used a voltage divider on the output to bring it down to a suitable level for the TRX chip clock input...
73, Oscar
Mit 0,1ppm ein sehr stabiler LO, allerdings benötigt er 3,3V, die im Pluto nicht vorhanden sind. Ich habe deswegen einen LowNoise Regler von Analog (passend zum Pluto
) verbaut.
73 de dd9fj / Mo
Hi Michael,
man kann die 3.3V z,B, auch über C165 auf der Platinenunterseite des Pluto beziehen.
Welcher TCXO Typ is das denn?
73, Oscar
Nur so eine Frage:
Wäre es nicht etwas einfacher gleich einen TCXO Typen zu nehmen, der CMOS als Ausgangswellenform hat ???
(bei manchen Typen gibt es ja Clipped Sinewave und CMOS Output Varianten)
73, Oscar
I am just in preparation of firing up and modifying my Nokia UMTS PA (the one with 2x BLF8G22LS-150) that I've got last year from the source in Poland mentioned here.
In order to get a decent PWR cable connection I just made small PCB adaptors for the 2x5 pin heders (2mm types)...
Since the price was identical for 4 pieces or 20 pieces I ordered 20 pieces - so I have many spare ones now
If someone is interested in a pair of these PCBs for their amp let me know by direct message.
73, Oscar
DK8KW OK Geri that makes sense and that must indeed work fine.
I was wondering if someone has found a way to use this software just stand alone without any additional local tone generator for local CW tone monitoring?
My money would be on non-optimal alignment.
How about the dish quality and or its feed mount position accuracy, if everything else has been tried already?
I am getting 3dB noise floor hump on a 75CM dish - that matches quite well other "well receiving" setups reported here.
How do you guys actually use this software?
I briefly tried it and deleted it again because the latency was making the software absolutely unusable for me...
When I was using it straight on my PC (= hardly any latency) it was usable to locally generate CW in my headphone or loudspeaker.
Even around 20WPM iambic paddle mode with the mouse mode worked very well for me with nice sounding CW.
But as soon as I switch the output to virtual audio cable to feed into SDR-Console for TX, the delay via the satellite makes it unusebale.
I therefore tried to activate the local echo output on my standard soundcard in the virtual audio cable settings for monitoring my CW locally, but even that had a terrible latency and unclean output of the monitor signal.
It would be actually great if it would heve two selectable outputs - one into a VAC for TX and one to the local soundcard (the latter with very low latency)...
73, Oscar
SWL - markro92: When I try to receive the beacon (with version .16 and Pluto) I manage to get a stable and nice constellation, but as soon as LDPC worker is activitaed the constellation frequently loses lock and nothing is streamed therefore to VLC.
Could this be a CPU issue? (its a i5 3rd generation)
Thank you Marcel. That sounds like great improvements in the pipeline. Do you think that by ading the 16 bit processing path you may have slightly less quantisation noise which could potentially lead to a slightly improved RX performance with SDRs >8bit....???
Or am I misinterpreting something?
73, Oscar
I had my equipment running over the last 24h for this experiment.
My hardware was a GPSDO disciplined Pluto SDR as RX for the IF.
And a LNB also GPSDO disciplined.
GPSDO was a dual channel Leo Bodnar unit programmed for 25MHz (for the LNB) and 40MHz (for Pluto) outputs.
The SDR-Console measured peak-to-peak drift was +/-26Hz over 24h.
See attached EXCEL graph below.
73 de Oscar
I had my equipment running over the last 24h for this experiment.
My hardware was a GPSDO disciplined Pluto SDR as RX for the IF.
And a LNB also GPSDO disciplined.
GPSDO was a dual channel Leo Bodnar unit programmed for 25MHz (for the LNB) and 40MHz (for Pluto) outputs.
The SDR-Console measured peak-to-peak drift was +/-26Hz over 24h.
See attached EXCEL graph below.
73 de Oscar
SWL - markro92 : Maybe the author of the Minitioune Software, Jean-Pierre F6DZP could help?
SWL - markro92 Are there any plans to intoduce somewhere in the GUI a simple MER display ? (Modulation Error Rate in dB)
I think that whould be very useful for people that are tweaking their hardware, RX, antennas, LNB etc. to "measure" some kind of relative improvements or changes. Similar to the hams always giving MER reports when using Minitouner hardware...
Maybe I missed it, but is there a feature in SDR-Console to write this data from the beacon lock window over time to file?
73, Oscar
For those of you whom my modification is too complicated and just looking for a "classic" way of using this Twin LNB only as a single LNB with externally fed 25MHz reference via one of the spare F-Connectors, here is a nice page by EA3CNO explaining how this could be done for exatly this same LNB type. (its in Spanish, but you can use Google Translator and/or the photos are rather self explainatory)
https://ea3cno.wordpress.com/2…b-con-referencia-externa/
73, Oscar
Last weekend I modified the Octagon Twin LNB Green. The main problem is that the version with the two cristals that the original poster added here is no longer available. Most of the versions sold today use only one central Xtal for both PLL sections of the LNB.
Since the idea was to use this LNB for DATV and for NB I modified it to have one output working with the internal Xtal reference for low phase noise (important for DATV on the WB transponder). At the same time I wanted to be flexible regarding the connected DATV RX and therefore it was desireable to have an output in a frequency range able to be received by am ordinary SAT TV RX.
On the other hand I wanted to have good stability by means of external stabilisation for the NB transponder.
Based on the good experience with my VENTON LNB single cable inyection solution (invented and heavily promoted by Remco PA3FYM) my target was to have the same advantage with this LNB.
So I had to come up with a creative solution to cover both use cases - optimized for WB and for NB at the same time....
Here is what I did:
The one side of the LNB is running the PLL chip on a 24 MHz Xtal (original 25MHz Xtal was removed with hot air gun).
This results in an IF frequency for the WB transponder of around 1130Mhz...(+/- Xtal tolerance)
To avoid the other PLL being fed also by the 24MHz reference part of the Xtal PCB traces were cutted and partially removed.
(see photo)
The rest of the circuit is the typical PA3FYM approach. Cut the power trace from the F connector to the meander line and 330 ohms resistor leading to the voltage regulator. A 120pf+330nH parallel circuit blocks the 25MHz reference but lets the current pass to the voltage regulator to power the future "NB transponder" section of the LNB.
Then only add a 120pF + 330nH LC resonant circuit as a filter for the 25MHz reference and use a short enamelled wire towards the inyection point of the PLL dedicated for the NB transponder....
That"s it...now one side works for NB (with 25MHz reference injected in the shack via one of the many diplexer / triplexer circuits found here on the forum). And the other side works well for good quality DATV reception.
Hope this will be useful for you.
73, Oscar
