I think the recent fix for this PN export issue still has a problem:

The frequency offset values in first column of the exported CSV are all a factor of 10e4 too small if the sweep is from 1kHz -> 1MHz.
If the sweep is from 100Hz -> 1MHz they are all a factor of 10e5 too small.
If the sweep is from 10Hz -> 1MHz they are all a factor of 10e6 too small.

Attachments:

You must be logged in to view attached files.

Thanks for the insights, much appreciated.

I grabbed the old API, and will have a play with it – minor issue was the Legacy API manual is linked to the Headless API manual on the SH.com website. The SH.eu website had a copy though.

SM200A – one day 🙂

Thanks Andrew, I see the latest update has the fix 🙂

One minor suggestion for later when the SM200A dust settles – how about changing the default colour scheme to dark, and the trace to green on black?
It’s just a sexier look than the default white and grey and a better first impression… (perhaps do it before you send the SM200’s out for review!)
Check out the images on the SH facebook page, that is how many people set it up.

Never tried with Bootcamp but I expect that would work just the same as on any other PC.
Justin might have been referring to using it under a VM such as Parallels, which I’ve done ages ago, but there were some issues with disconnects.
One improvement which would be helpful is a single click re-connect – i.e. toolbar button. Currently reconnecting takes three clicks which is a tad painful.

You guys had been very quiet for a while so I was hoping something was being cooked up 🙂

Spec sheet is up here in case anyone missed it:
https://signalhound.com/ims-2017/

Phase noise spec looks good.

Are you able to go into any more detail about that process Justin?
Is the ADC actually running at 24bits or something less?
What is inside the ‘digital converter’ process?

Thanks,
Roger

Sorry for confusing you catshot. All that info was actually in my post, but you wouldn’t be the first person who didn’t understand what the hell I was on about 🙂

Cheers,
Roger

Hand sweeping continuously is not so good, as the software will detect the changing signal as spur and remove it. So you need to pause your sweep regularly to let the software determine the signal is valid.
Also you could turn off the spur detection – of course you then may have issues with spurious and images appearing.

Max hold is available, which would avoid having to capture the trace and post-process it. With spur detect turned on it takes more patience to ‘fill in’ the trace though.

Just upgraded and the sweep recorder/freq mask trigger is a bit annoying with the drag resize and buttons. That doesn’t match the way the other parts of spike work.

I think it should have a minimise icon like all of the grouped settings on either side of the sweep display.

Put the two headings on the same line and remove the buttons to save vertical space. Just display/hide the pane when the heading is clicked. (Sweep recorder or Frequency Mask Trigger).

Hi Mervin,
I have had similar problems, it means that the 10MHz out from the TG isn’t coupled to the 10MHz in to the SA.
I don’t have it in front of me but I had problems last time I tried to get this to work (on Win 7). The tick boxes which are supposed to show the status of the 10MHz I/O don’t. 🙂
So you can probably fix it by making sure you have the 10MHz connected, and fiddling around with the 10Mhz reference settings.

Hi Phil,
If you have time, could you advise the spec of your KCDSR30-31?
I can’t find any info about it on the net unfortunately.

Thanks,
Roger

[hendorog]

Hi Justin,
Yes something like that would work. I use it to remove offsets which are not uniform. E.g. the IF output of my HF radio does not have a flat noise floor and I could flatten it by removing a saved trace done with the antenna disconnected.

I though it could be added as extra options in the trace type combo – subtract trace x or add trace x, and then you need another combo to select ‘x’.

If you added another trace type called ‘constant’ then you can do trace 2 – trace 1 – offset, by linking the traces together.

Cheers,
Roger

• This reply was modified 10 years, 2 months ago by hendorog.

Hi AJ,
Yes I understand that you don’t want to support the many VM implementations around.
Regarding the old software in the VM – it worked fine without the suppress error option turned on. That option did help however when the VM was minimised – i.e. when I was using a native application. Presumably there were less resources dedicated to the VM.
The difference is this; In the old software, when a comms error occurred the connection to the SA was recovered automatically. Whereas the new software disconnects when there is an error and needs to be reconnected from the File menu.
In a nutshell – I’d rather give you guys my $$ for a TG than give it to someone else for a dedicated Windows machine for this 🙂

Cheers,
Rog

Hi guys,
First, thanks for all of your hard work. Andy hit the nail on the head just above – its been a revelation seeing how polished this software is, and how the hardware can really perform.

Second, I’ve found a few issues and have a couple of suggestions which I better write down before I forget:
* The spectrogram on/off state isn’t restored in the preset.
* Realtime mode limits the Span to 100kHz min. But this limit can be bypassed by going to Sweep mode, reducing the Span, and then selecting Realtime mode.
* Spike crashed on me a couple of times when switching from Real time to Zero Span, but I can’ reproduce that now.
* I think that the 3-D spectrogram should interpolate samples to form a ‘continuous wall’ instead of a ‘row of spikes’
* I’d like to be able to subtract and add traces to/from each other, which could be done in the old software
* I used the old software on a Parallels VM on a Mac and it worked fine – although sometimes the Suppress Errors helped when switching in and out of the VM .
In the same environment Spike starts up and appears to work, but dies part way through a scan with a Device Connectivity Error. I can’t find the Suppress Errors option anywhere. So its not usable for me in that environment.

Cheers,
Roger

[Author]

Posts