I’ll be willing to give that a try this evening on my V1.2 board and see if that makes the improvement - stay tuned,
73 de Marty, KN0CK
1 Like
I have a V1.2 and V1.4 board and I will personally look into the changes that @sdr_research has suggested to see if that indeed fixes the HF issue. That won’t be for another 8 hours or so because my work day just started here. Stay tuned for more later -
73 de Marty, KN0CK
@andrewback Thank You for the updates Andrew. After much thought and based on your comments, including that your company feel that it is user error, I am going to take Crowd Supply up on their offer of a full refund. This will obviously not be of any negative effect on your company I am sure as even if a few hundred people receive refunds then you will still be above your goal. I will continue to watch the threads and still may well decide to buy a used Lime.
Wow i’m glad I came to these forums… my limeSDR was due to ship next week… have contacted crowd supply to see what my options are. If its useless for HAM radio (as advertised on crowd supply) then theres little point me keeping it and I might aswell get a refund. Was really looking forward to getting it too!
Jamie
M6XRJ
Jamie, This is the option that Crowd Supply have been kindly offering.
I was in touch with the team earlier today about this very issue and
they are looking into the reported change in frequency range. Their
initial guess that the report is incorrect due to some sort of set up
error and they intend to update all backers with their findings soon.
They only found out about this report yesterday and are working
quickly to get to the bottom of it. Your order is likely to be shipped
in by the end of next week, if not this week. If you’d like, we can
hold your order until the LimeSDR team posts their update. Of course,
it’s not a problem to cancel your order now as well. Please advise.
@N2EME just to be clear, if a difference in performance between v1.2 and v1.4 hardware is observed, all we can assume given our testing is some form of misconfiguration and/or inaccuracy in the tests, or failing that an issue with the hardware. There is no suggestion that sub-optimal HF performance is down to user error — this is due to the fact that the matching networks were selected to provide optimal performance at higher frequencies.
Ok, when exactly did this escalate into being useless for ham radio?
The issue observed concerns poor performance on HF, but hamradio is much more than HF. Personally, I got the board for experimenting on the 1.2, 2.4 and 3.4 GHz bands, where we have very large allocations and very few users 
4 Likes
Sorry, should have been more specific- for HF ham radio. which for me personally is all I have an interest in anyway.
@andrewback Hi Andrew, I can understand what a difficult situation you have found yourself in and as you have most certainly have noticed I have been very positive in my comments and allowed others to interpret the information that you have kindly provided.I think though if you re-read your comments you may find them a, shall we say, a little different from what you intended to write. My own personal interpretation of what you said was that the software writer did not know how to use the software that he wrote. It came over very much as a HRD comment although I am very sure that it was not taken in that way. That, and you company saying that they were not aware of a problem until the 8th Feb despite being informed on the 23rd Jan of the same problem, was what made me choose the route which I reluctantly did.
I had originally looked at the Lime for two reasons, for 144MHz EME RX (and above) and also in real hope that you would produce a future version that would cover 10GHz. I will therefore continue to wish both yourself an Lime well and continue to monitor in hope that both of these can come to fruition.
I do have one very simple question, purely out of interest although quite honestly your reply to which may well get many people back on board. Many Hams do a simple test, lets call it a noise floor test as that is how they will understand it. If you apply a signal from a Signal Generator at what point does the signal disappear from SDR Consoles waterfall. I will stick with the SDR console waterfall based on you saying your engineers were testing with it. I was wondering if you could publish the -dBm figures for say 3.5Mhz, 7MHz, 28Mhz, 50Mhz and 144Mhz. Just a very quick down and dirty test but will tell a lot of hams a number which they can relate to in their head.
Again, Andrew, you have been most helpful in getting the information out on the Lime and for that I most certainly thank you.
Paul
1 Like
Ok, no problem 
If the intended use is limited to HF, then this board would not be my first choice, even if it lived up to what was advertised.
1 Like
Apologies if this is the case and it certainly was not the intention. To expand on which: there have been changes in the way that v1.4 hardware is programmed when compared to v1.2. Some of the issues which arose from this and have since been resolved turned out to be in the driver itself, whereas others have been in the way that it has been used to configure the hardware. So, misconfiguration (of the LMS7002M) could be in one of two places.
If a difference in performance is observed between v1.2 and v1.4 hardware, this is something we want to get to the bottom of and are confident in can be addressed in software.
Your positive contributions are noted and have been welcome, Paul
More to come…
My advice to fellow HAMs is to keep the ball down. Here reading some posts which are near panic level. Even if the board is performing good above 10MHz or 100MHz, things can be nice translated on lower frequency. Please keep in mind that we agreed to be part of development, that can be sometime bumpy road 
My first board is about to be delivered to UK address next week, custom paid.
So hope can publish soon some of my own measurements…
73 Djani
PS looking for second LimeSDR…in case any of you may have fair offer, pls find my email on QRZ.com
3 Likes
Make LimeSDR Great (Again?)!!!
I am personally concerned that are software bugs and hardware bugs. If the matching network behind the wideband transformer shorts the differential signal, which has been implied, then no amount of driver work will solve the problem. If there is a problem with the driver as well, then that problem is a different issue entirely.
I wish the problem would just ‘‘go away’’, but that is not how the world works. And, I think that many people may have the impression that HF is difficult to receive which means the problem may have been masked, because those who had the hardware had only subjective ideas. But, does anyone have any real screenshots from the HF reception?
If the hardware works, let’s see it, If the hardware doesn’t work, Let’s see it too.
Show me the reception!!
I have been asking on IRC for over a year for decent screenshots.
Goatman
2 Likes
Hi All, a further update and some recommendations for optimising performance on HF:
1 Like
https://myriadrf.org/blog/optimising-limesdr-matching-hf/
@andrewback so you are saying that we bought the Lime as advertised, but now we have to modify it ourselves to fix an issue that should have been fixed before it ever shipped…
3 Likes
Noticeable improvement !
But,the HF is still 40-60dB down comparing to the 100Mhz, the LF bellow 1MHz , forget about it 80dB down. I agree that there is a hardware (LMS7002M) limitation and we should live with that, no free lunch.
I think that HF guys will not be so happy, they can start building the converters…
Optimising LimeSDR matching for HF
By Danny Webster
10th February, 2017
Figure 2 shows the Return Loss and VSWR of the Broadband network from 20MHz to 500MHz. The matching network forms a terminated differential transmission line, and has good performance from 30kHz to 400MHz.
Dear Andrews,
In the report of Danny Webster 10th February, 2017
You should read 30 KHz? Or a typing error? and read 30 MHz to 400 MHz?
Bernard MALET
Looks great!! Thanks for the update!! The 100nf capacitors should be an easy install. Looking forward to purchase!
Danke!
Goatman
Could you link the most recent schematic for Lime …
I have LMS7002EVBv3.1 and i don’t see C14 (dogh, found it – search tool didn’t)… though i can understand from what i have for a schematic…
Even though the TC1-1-13MA is not rated @ 1Mhz for insertion loss, im sure its with in 1db of BCR-221JLB. My problem is that i cant tell from the TC1-1-13MA PDF if its a 1:1 or 1:2 ( think its 1:1), never the less, the BCR-221JLB has an insertion loss of .5db compared to TC1-1-13MA’s .18db(@ 4.5Mhz) …
This quote from the optimising-limesdr-matching-hf link, is unfortunate but not crippling. Hams need to remember that signals on HF are usually large compared to microwave bands … and local noise is bad.
Hello Bernad,
You are correct. It should read 30 MHz, not 30 kHz.
Karolis
1 Like