|
I own two TinySA Ultra devices. I am attempting to measure the output level of a frequency synthesizer and it appears that one of the TinySA Ultras is giving incorrect levels. As an example, this morning I measured the output of the synthesizer at 1246 MHz on each of the TinySA Ultras. One device showed a level of -6.2 dBm and the other showed a level of +3.2 dBm. The +3.2 dBm value is close to the synthesizer's specifications. I did a level calibration and a self test on each TinySA Ultra just prior to making the measurements. The same cable was used for both measurements and I have also noted the discrepancy with other cables. I have a male to female adapters on each TinySA Ultra to protect the device inputs from wear and tear and also got the same reading when I removed the adapter. Both devices are running firmware version 1.4-78. Am I forgetting something or is one of the TinySA Ultra devices defective? Thanks for any suggestions. --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
As an aside, I did take measurements with both TinySA Ultras with a 30 dB attenuator inline and the results differed by the same amount...almost 10 dB difference between the two devices! --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Alright, after some more testing I am even more confused! I tested the two TinySA Ultras with a Siglent signal generator at 50 MHz and the two TinySA Ultras were within 0.3 dB of each other and consistent with the level set on the signal generator of -20 dBm. When I connect the TinySA Ultras to the Q5 Signal synthesizer, the level readings still differ by about 10 dB. This points to some defect in the synthesizer, but why would the two TinySA Ultras cause its output power to differ by 10 dB??? If anyone can shed some light on this issue, please let? me know. --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
What kind of cable are you using? ?I have had issues with Pasternak cables at 2GHz.
|
Charles (I hope that you go by that name...I looked you up in QRZ.com), I have tried cables that I received with my TinySA Ultras and NanoVNA devices and all seem to give the same results. I always use the same cable when I am comparing results from the two Ultras. --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Can you run a selftest on both devices and capture and post the end screen of the selftest where it shows the result of test 13? This should give an indication if the level calibration of the two devices is different. And can you do this test on both devices to rule out any attenuator issues:? Further it seems the issue appears only with higher frequencies, can you check if up to 699MHz (or the highest frequency you can generate below 699MHz) they have a? similar performance and they are different at 801MHz (or the lowest frequency you can generate above 801MHz)? --
For more info on the tinySA go to https://tinysa.org/wiki/
|
Working on it! --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Erik,
I have made the measurements that you suggested and have attached the files. I could only generate a lower frequency of 435.1 MHz (closest to 699 MHz) and a higher frequency of 823.0 MHz (closest to 801 MHz) with my synthesizer.
Your analysis would be appreciated!
Ed, K0KC
k0kc@...
http://k0kc.us/ --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Let me know of you disagree with any of these conclusions
Last selftest: Both tinySA have same signal to noise at 30MHz with LNA on?
Attenuator test: Both tinySA have a good attenuator
400MHz test: same noise and same signal level for both tinySA
823MHz test: same noise and same signal level for both tinySA.
Conclusion
I do not see a difference between these two tinySA What is happening? --
For more info on the tinySA go to https://tinysa.org/wiki/
|
It appears that the problem may be the synthesizer, although I do not understand why its output level changes when I switch TinySA Ultra devices. I just recently purchased this synthesizer and will discuss these findings with the designer/seller of this device at Q5 Signal. He is sending me another synthesizer which I should receive tomorrow. Perhaps the new synthesizer will behave differently. --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Just a thought; you are measuring an o/p of +3dBm, is the input protection on one of the TinySAs cutting in a shade before the other??
73
Jeff G8HUL
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] < [email protected]> On Behalf Of Ed Wilson via groups.io Sent: 10 April 2023 18:09 To: [email protected]Subject: Re: [tinysa] Level Discrepancy Between 2 TinySA Ultras It appears that the problem may be the synthesizer, although I do not understand why its output level changes when I switch TinySA Ultra devices. I just recently purchased this synthesizer and will discuss these findings with the designer/seller of this device at Q5 Signal. He is sending me another synthesizer which I should receive tomorrow. Perhaps the new synthesizer will behave differently. -- Ed, K0KC k0kc@...
|
I received a replacement synthesizer from Q5 Signal today and ran some tests which definitely showed that the two TinySA Ultra devices respond differently at the higher frequencies. I made sure that the two TinySA Ultras had exactly the same settings and I utilized the same cable for all measurements. The comparison follows:
Since both synthesizers give similar results, I will be returning one of them to Q5 Signal and assume that the issue is with Ultra 1 at the higher frequencies (Ultra 2's results are consistent with the synthesizer specs). I believe that I bought Ultra 1 from the Zeenko Store and Ultra 2 from R&L Electronics. R&L did not place a serial number on the invoice and I have not yet tried to locate an invoice from the Zeenko Store to see if they included a serial number. I have had these devices for several months now and I am not even sure that a warranty would be in effect. It is unfortunate that I was not able to make this comparison in a more timely manner, but I just did not have an appropriate signal source.
Ed, K0KC
k0kc@...
http://k0kc.us/ --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Thanks for the idea, Jeff.
I did some tests with an attenuator in line to reduce the input signal level and still observed the discrepancy.
Ed, K0KC
k0kc@...
http://k0kc.us/ --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
On Tue, Apr 11, 2023 at 10:05 AM, Ed Wilson wrote:
I did some tests with an attenuator in line to reduce the input signal level and still observed the discrepancy.
?
Ed, ? If you don't mind, I'd be interested whether you noticed the same discrepancy if you used one Ultra in the signal generator mode and the other in the spectrum analyzer mode and then reversed the set-up.? ?If the same discrepancy is still apparent, then it would be interesting to get feedback from other users with two Ultras who notice the same issue. Herb
|
On Tue, Apr 11, 2023 at 12:20 PM, hwalker wrote:
..If you don't mind, I'd be interested whether you noticed the same discrepancy if you used one Ultra in the signal generator mode and the other in the spectrum analyzer mode and then reversed the set-up.?
?If the same discrepancy is still apparent, then it would be interesting to get feedback from other users with two Ultras who notice the same issue.
BTW, ? The generic level correction tables were based on the assumption that the manufacturing tolerance would be within ~+/-2dB for all devices and preproduction testing on a small sample of units was within expectations.? It could be that some component differences in a larger production sample size cause the generic level correction tables to show the discrepancies you are noticing. ? One of the exciting things about the Ultra is that Erik provides the user the means of editing all the correction tables to get more accurate results than the generic tables, if they have the equipment to do so.? In your case, if one of your Ultras is proving to be more accurate than the other than you can use it as a source for correcting the level correction table of the other. Herb
|
Herb, Thanks for the idea! I will try to find time to do this later today or if not today, very soon. --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
Herb, I do not recall seeing information on editing the correction tables in the Wiki...can you point me to some documentation on the procedure? --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
On Wed, Apr 12, 2023 at 03:48 AM, Ed Wilson wrote:
I do not recall seeing information on editing the correction tables in the Wiki...can you point me to some documentation on the procedure?
Ed, ? If you type the command "correction" with the Ultra connected via a serial terminal (tinySA-App, PuTTY, etc) you will get the following response:
correction
usage: correction low|lna|ultra|ultra_lna|direct|direct_lna|out|out_direct|out_adf|out_ultra|off|on 0-19 frequency(Hz) value(dB)Those are all the correction tables used by the Ultra and which are editable by the user.? The tables are for both input (specan) and output (sigen).? Erik and Kurt Poulsen developed each of the tables and are the "goto" experts on making changes to a particular table.? I don't recall either of them creating a document or video describing the process of editing a table but several users have reported doing so with success. Your discrepancy appears to be in the Ultra range (>800MHz) so that is the table you will probably need to edit.? If you type "correction ultra" you will get the following response, which are the generic correction table values for the ultra mode:
correction ultra
index frequency value
correction ultra 0 30000000 0.0
correction ultra 1 700000000 0.6
correction ultra 2 980000000 1.7
correction ultra 3 1440000000 4.5
correction ultra 4 1590000000 4.5
correction ultra 5 1900000000 3.2
correction ultra 6 2810000000 4.6
correction ultra 7 3340000000 6.3
correction ultra 8 3390000000 5.7
correction ultra 9 3930000000 7.0
correction ultra 10 4230000000 8.8
correction ultra 11 4300000000 7.0
correction ultra 12 4340000000 8.3
correction ultra 13 4810000000 11.4
correction ultra 14 5070000000 11.6
correction ultra 15 5110000000 13.3
correction ultra 16 5300000000 12.4
correction ultra 17 5510000000 12.6
correction ultra 18 5850000000 15.8
correction ultra 19 6000000000 15.9 I believe the format for editing a table entry is: correction ultra index(0-19) frequency(Hz) value(dB)I also believe you have to send a "saveconfig" command after the completion of all entries to make the changes "stick".? If I recall correctly, the entry with value 0(30000000Hz) should not be edited. Rather than using a serial terminal to make changes to the tables another method is to use the Ultra itself.? Select Config>More>Level Correction and the following screen will appear allowing the user to select a table for editing:  Perhaps one of the users who has actually edited one of the Ultra's tables can start a separate topic and document their procedure for you. Herb
|
Thanks, Herb! I will wait to see of anyone with experience chimes in before I attempt any edits. Am I correct in assuming that I need to increase the "generic" value in dB if I want the display value to increase at a particular frequency? --
Ed, K0KC
?
k0kc@...
http://k0kc.us/
|
On Wed, Apr 12, 2023 at 07:26 AM, Ed Wilson wrote:
Am I correct in assuming that I need to increase the "generic" value in dB if I want the display value to increase at a particular frequency?
Ed, ? I'm sure Erik or Kurt will correct me if I'm wrong, but from the way the table values trend upwards at the high end of the frequency range, it looks like the "generic" values in dB are added to the raw measured values to correct for loss.? So, if the displayed value is low, then as you said, you would need to increase the generic value in dB by the amount of the displayed value error. Herb
|
W5CDT
