@SimonG4ELI On which RX input is it 15 - 20 dB down compared to Airspy (in FM band)? I would have thought that on RXLNA-W it shouldn’t be that much filtered. And 5 - 7 dB noise figure isn’t that bad (Airspy has 3.5 dB noise figure if I remember right).
If you look at the circuit schematics provided by this link
You will see that the so called " wideband" LNAW input is stated as only having a frequency passband of 700MHz - 2.6 GHz .
The LNAH input is defined as 2- 2.6 GHz and the LNAL input as 700- 900MHz
This supports our concerns about wideband performance options . The filter component values are also supplied and we can use these to estimate the filter characteristics using a suitable RF network modelling software .
- RX L: 700 MHz - 900 MHz
- RX H: 2 GHz - 2.6 GHz
- RX W: 700 MHz - 2.6 GHz
- TX A: 30 MHz - 1.8 GHz
- TX B: 1.9 GHz - 2.6 GHz
So what is right? Can you please give us plots for all RX inputs and TX outputs @RicardasVadoklis or @andrewback? So we don’t have to speculate (or waste time with modeling / simulation) and can know exact characteristics of filters.
This is a serious issue for a lot of funders. Most of us cannot modify SMD Designs and bought the board to use out of the box. So please lime find a solution for good. There’s still time enogh to do it.
And please: don’t say it was written in the schematics. You did not adress engineers but a broad audience. You might as well have published the specs in kisuaheli (no offense to people speaking that language intended).
Maybe we should create another topic to discuss this issue? This one at first sight is for RX and TX connectors which might be confusing.
Currently we are using components on LNA inputs to fine tune for some cellular bands. On the final version of the LimeSDR we shall match the receivers for the optimum broadband performance as shown in the pictures below:
This is great news @RicardasVadoklis, thanks for the update!
Many thanks for this confirmation . I truely believe you have made the right design change here to widen the market potential for this board and honour the way the product has been marketed . The performance plots look good . I look forward to receiving my boards and testing them in a number of applications I have in mind both professionally and for hobby experimental use.
David Price CEng MIET , PA4MSA
Thank you very much, @RicardasVadoklis. Any idea what the response will be below 100 MHz?
@RicardasVadoklis Thank you, this is excellent to know! Btw. can you please also publish similar plots for RX LNAW, TXout1 and TXout2?
All, apologies! Hadn’t appreciated that the matching networks would be tweaked for the campaign boards!
Also are we now all in agreement that these are matching networks and not filters? As such and like matching networks on other platforms, they will obviously have areas where they perform best, but filtering is left to off-board devices — which as far as I know is something that matching cannot be offloaded to. That is, matching TX and RX to a 50 ohm line and not to the antenna (which ATUs are used for, of course).
More info for limesdr board after 1:47:00
Just a reminder what Lime Micro promised to offer as the specification in terms of broadband gain and NF for the 1.4 units . It seems that this has not been honoured with their design choice of matching network components at the antenna port inputs
Please can Lime Micro comment here
David Price , CEng MIET
I am a backer and I want a comment too.
@PA4MSA @freebil I understand that, following overall feedback from experiences with v1.2 hardware, the decision was made to not change the matching networks. I appreciate that this will not be welcome news for some, but we are looking at a number of solutions and a detailed update with options will be posted later today.
I greatly appreciate the posting by Ricardas of plots for the RXn_L and RXn_H antenna ports. I would like to see a similar plot for the RXn_W port. I would also like to see a similar set of plots for the TXn_1 and TXn_2 output ports, so that I could make a maximally intelligent choice of which of the 10 Rx/Tx ports to use for a variety of applications.
Once again, thank you very much for the work done so far.
John Toscano, W0JT/5, EL09vu20
That is very bad news.
It was promised to have a broadband matching network instead of the very narrow UMTS/LTE optimized one.
Please post detailed analysis of how the final HW and the “HF fixed” HW behave when a signal is sweeped on the 3 RX antenna ports from 100 kHz to 3.8 GHz.
The interesting thing is the behaviour from the antenna input plug to the final I/Q output on the computer.
And then sweeping a signal on the 2 TX ports and look at the output signal on the antennas.
These are the optimized frequencies for the current matching network:
RX1 L = 700 MHz - 900 MHz
RX1 H = 2 GHz – 2,6 GHz
RX1 W = 700 MHz – 2.6 GHz
What does that mean.
How much attenuation do I see if I am in a non optimized frequency?
RX1L is optimized for 700-900 MHz (veeeery small frequency band)
What behaviour will I see if I go outside of that optimal band?
0.95% or 0.05% of the signal that would have made it in the optimized band? something in between?
How does the “HF Fix” worsen the behavior on higher frequencies?
We need information that we can decide weather to cancel the purchase, or go for the HF fix or go for unmodified HW v1.4.
Please give detailed answers, as we need a basis for the decision if we want to cancel or not.
Out of curiosity what were the motives to break the promise to change the matching network to be much more broadband??
How does the frequency sensitivity compare to Ettus USRP and the BladeRF and HackRF.
They use 1 matching network for the entire frequency spectrum they support. How does their sensitivity on certain bands compare to the LimeSDR’s? (when taking into the account everything from the antenna plug on)
Personally I find it sad that we have not been informed earlier that the promise of a more broadband matching network has been broken.
I think all your questions are answered on this forum already, multiple times. And for those specifics, you can find in the documentation. As for comparing the other SDRs you can look at the docs and find that info and there was a table with some of those comparisons on the crowdsupply page.
No promises were broken and you would know if you read through just a few of the posts here. There was a lot of discussion also with beta testers that were interested in HF last year. So you may want to contact those testers.
And a lot of the other HF fans that were threatening to cancel have changed their minds.
I was refering to this promise made in this very thread:
“Currently we are using components on LNA inputs to fine tune for some cellular bands. On the final version of the LimeSDR we shall match the receivers for the optimum broadband performance”
The final LimeSDR hardware is still optimized for cellular network bands, and was not made more broadband.
Anyhow … at the moment I think I will do nothing, and when I have the HW I can still decide if I want to remove MN18 or leave it as it is.