Segmented Sweep

[KD5JHA]

I read a thread about how to do this a while back and it seemed to be pretty straightforward but when I went to try to increase the number of points in the latest v1.6 rev software it will not let you do so. Has this capability been altered or removed? I am also having some issues with the TDR function properly displaying data so not sure if there are some bugs in the new release or if I am just not operating it properly. I am new to the platform so excuse me if I am just making a simple mistake.

[jankae]

Segmented sweep: this should be as simple as changing the number of points to something larger than 4501 (which is the maximum a single segment can do). Just make sure you have enabled the segmented sweep option in the preferences:

I just tested this with the latest development version and it works. The latest release (v1.6.2) also works and so should earlier versions (I have not changed anything in this part of the code for a while.

I am also having some issues with the TDR function properly displaying data so not sure if there are some bugs in the new release or if I am just not operating it properly.

If you can tell me more about what is not working properly, I might be able to tell you if this is a bug.

[KD5JHA]

I found the correct place to enter the information for the segmented sweep. I just was trying to do it from the wrong menu. Like I figured, just a rookie mistake. I will take another run at the TDR issue and see if I can get that to work now and if not I will take some screenshots and sent it to you.

…

On Tue, May 6, 2025 at 12:21 AM Jan Käberich ***@***.***> wrote: Segmented sweep: this should be as simple as changing the number of points to something larger than 4501 (which is the maximum a single segment can do). Just make sure you have enabled the segmented sweep option in the preferences: image.png (view on web) <https://github.com/user-attachments/assets/a54b88f7-bccd-491a-bc0e-dfb9e54a4b71> I just tested this with the latest development version and it works. The latest release (v1.6.2) also works and so should earlier versions (I have not changed anything in this part of the code for a while. I am also having some issues with the TDR function properly displaying data so not sure if there are some bugs in the new release or if I am just not operating it properly. If you can tell me more about what is not working properly, I might be able to tell you if this is a bug. — Reply to this email directly, view it on GitHub <#310 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BSFQWMZUZ5UDHTAXAMJW3QD25BBFLAVCNFSM6AAAAAB4PSLLR6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTGMBUGY2DSNA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- R. Matt Gholston

[KD5JHA]

Ok,

I had a chance to play with the unit a lot more yesterday and sweep some longer transmission lines and antennas. It worked well and gave me some clean data. I am still having some issue getting the TDR/DTF feature to work in real time in the GUI but I am confident that it is me and not the software. Probably the order of operations that I am doing things.

I can dump the data to an S1P and load it into another viewer program that I have and see it just fine in the time domain.

I went on the LH side of the screen in the traces column and created a new trace called DFT that is a math operation.

I tried using the Time Domain Gating Setup that you have preconfigured and it pops up a window where it shows me some data where I can pick the gate window area but after I hit OK when I return to the main screen and assign one of the X/Y graphs to view this parameter it is just blank. No matter if I assign the setup to Time,Distance or Frequency it is blank.

I am confident I am making a simple error so if you might give me a step by step through how to set this up...

When I have used this type of feature (Distance to Fault/TDR) on other devices Either a Dedicated TDR or FDR or a VNA in the past I am used to looking at RL or VSWR on the Y axis and Feet or Meters on the X Axis. Sometimes I will look at time in pS or nS but not as often.

I am fully confident once I figure out where all of the things are located in this gui and what your thoughts are on how to make things work everything will make sense. But even reading the manual and searching online I still feel like I am tripping over myself trying to set everything up and then when you close it everything is gone and you have to set it all up again the next time. Can you save an often used gui setup so you can reload it later? That would sure be nice?

Another couple of things that would be nice just from an order of operations standpoint would be when you do a manual calibration using a regular mechanical OSLT kit if the software would automatically advance you through the steps and prompt you for the next stage so you don't accidently sweep the incorrect standard over again. Just a small optimization.

One last question...

I love the LibreCal, really saves me a ton of time. How would I go about generating some modified data for it in order to be able to mount some adapters on it to change the connectors to Type N Female? I have a set of really High end Pomona Gold Plated adapters rated to 18 GHz that would work well but I would like to recharacterize it with those on it so I can use it like that. Is that something a cal lab can do for me and if so what type of file would I need to generate?

[jankae]

I am not entirely sure what you are doing, screenshots would probably help a lot. But let me try my best to help:

TDR:

I went on the LH side of the screen in the traces column and created a new trace called DFT that is a math operation.

There are two types of math operations:

1. "from math" traces. These are traces which are calculated based on other traces. For example if you want to see the difference between S21 and S12, you can create a "from math" trace. But that is not what you want here.
   

2. Math operations on traces. They apply math calculations to a single trace and are configured on the right side of the trace edit dialog. This is where you want to configure your TDR operation:
   

I tried using the Time Domain Gating Setup

Time domain gating is not TDR. Sure, it uses a TDR internally but that time domain data is then filtered (the gate) and transformed back into the frequency domain.

If you want to see time domain data, you need to use the "TDR" math operation:

The simplest way to get to a TDR graph (from the default configuration of the GUI) goes like this:

1. Create a new trace, pick a name and parameter. I have named it "TDR" here and am using S11:
   
   As you can see on the right side, this trace is outputting frequency domain data. But we want time domain data instead.
2. Add a TDR math operation to the trace. You can leave all settings at default for now:
   
3. Close the trace edit dialog. You now have a new trace with time domain data but it is not visible because it has not been added to any graph. The "M" in front of the trace name indicates that there is a math operation active on that trace:
   
4. Drag-and-drop that new trace on the bottom left graph (shows S21 by default). Any other XY-plot will of course also work, I am just choosing the bottom left one as an example. There will be a pop-up asking you whether you want to reconfigure the graph to support time domain data (a graph can only have one X axis which means it can not show frequency and time domain data at the same time):
   
5. Click yes. S21 is removed from the graph, the X axis configured to time and the new TDR trace is added. The graph probably looks quite strange at first because it defaults to show the magnitude of the impulse response over the whole unambiguous time range:
   
6. Configure the graph as desired. For example, I have switched it to show impedance vs. time here and zoomed in quite a bit on the X axis:
   
   This example is a measurement from an uncalibrated LibreVNA, so the result is bad (the port is actually open) but it should be sufficient to show TDR configuration.

Other questions:

when you close it everything is gone and you have to set it all up again the next time. Can you save an often used gui setup so you can reload it later? That would sure be nice?

Sure you can: File->Save Setup will store everything (except your preferences and the calibration) in a file which you can later load again by using File->Load Setup. The calibration is not stored on purpose to avoid accidentally loading an old calibration. But if you want to, you can also save/load the calibration under Calibration->Save and Calibration->Load.

Another couple of things that would be nice just from an order of operations standpoint would be when you do a manual calibration using a regular mechanical OSLT kit if the software would automatically advance you through the steps and prompt you for the next stage so you don't accidently sweep the incorrect standard over again. Just a small optimization.

That is on purpose. I generally try to design everything for flexibility. This also means that the learning curve is a bit steeper because the GUI does not guide you very well through processes. But I prefer the advantages it gives you. In this case the flexible calibration system allows you to retake any calibration measurement as often as you would like and even after the calibration was already applied. The downside is that you need to pay more attention to click the correct measurement during the calibration, but I am willing to accept that.

I love the LibreCal, really saves me a ton of time. How would I go about generating some modified data for it in order to be able to mount some adapters on it to change the connectors to Type N Female? I have a set of really High end Pomona Gold Plated adapters rated to 18 GHz that would work well but I would like to recharacterize it with those on it so I can use it like that. Is that something a cal lab can do for me and if so what type of file would I need to generate?

You need touchstone files for all calibration standards. There are two approaches you can use here:

1. Attach your adapters to the LibreCAL and measure its behavior (with the adapters) for all standards. You need to measure it with each port set to open, short and load as well as all combinations of throughs. All in all this results in 12 s1p files and 6 s2p files. Name them accordingly (see manual for the LibreCAL) and place them in a folder on the writable LibreCAL mass storage device drive.
   • Advantage: very accurate results
   • Disadvantage: you need to take lots of measurements
2. Measure your adapter only (loss, delay and return loss if it is not very high) and use the result to calculate new touchstone files based on the LibreCAL factory calibration and the adapter data. Basically you combine the LibreCAL and the adapter mathematically.
   • Advantage: just a few measurements of the adapter
   • Disadvantage: you need to do some math and have to be confident that you do it correctly

For both approaches you need a good VNA with a way to calibrate to the reference plane of your adapter (Type N female calibration kit in your case). It is also very easy to make subtle mistakes in the calibration which will then be transferred to the new LibreCAL coefficients. Overall this is doable but certainly not the easiest project.

Also, I would be careful with the adapters. If you attach SMA-to-N adapters directly to the LibreCAL, the SMA connectors will be under a lot of stress when connecting cables to the N connectors. I believe it is quite likely that you will damage the SMA connectors over time unless you are very very careful with that setup.
If you want an eCal with N connectors, I would probably attach the N connectors to some other housing and use short cables to connect them to the LibreCAL ports. Of course you then need to make sure that the whole setup does not move anymore after it has been characterized.

[KD5JHA]

Thanks!