Since we have received multiple reports of damage to the attenuator due to input signals exceeding the limited power, starting with the serial number SA-20020001, we have printed the maximum power warning label on the surface.?Next?month,?after?the?front?panel?that?has?been?printed?with?silk-screened?port?labels?is?used?up,?the?port?labels?will?also?be?directly?printed?with?lasers.

I will continue to pay attention to community feedback and actively improve production.

On 2021-02-04, at 17:02, Hugen <hugen@...> wrote:

SA-20020001

Is that the first one from the batch that was made in the year 2002?

SCNR.

Grü?e, Carsten

Hard to read. Does it give max power in?
Ray

On Thu, Feb 4, 2021 at 08:36 AM, <rcbenedict1545@...> wrote:

Hard to read. Does it give max power in?

Ray,
? The max power is printed on the front in a format that is standard for commercial spectrum analyzers, " +10dBm, 10Vdc".? +10dBm RF or 10Vdc are both maximum limits.

- Herb

On 2021-02-04, at 17:56, hwalker <herbwalker2476@...> wrote:

+10dBm RF

If you know the inputs are 50 Ω, yes.

Then 10 dBm is equivalent to

0.707 V (rms)
1.000 V (peak for sine input)
2.000 V (peak-peak for sine input)

(It would be nice to know which of these are the real limits — it says “power” so it would be the rms voltage, and there would be no problem with peaks higher than those exhibited by sine waves.)

Grü?e, Carsten

Anyone who would assume the input not to be 50 ohms better not play with RF test equipment and go back playing in his sand castle...?

On Thu, Feb 4, 2021 at 08:02 AM, Hugen wrote:

I will continue to pay attention to community feedback and actively improve production.

Hugen,

I have a suggestion for you.? Why not add one of these SMA 20 dB attenuators to the TinySA package?? They are available in China for only a few dollars and would only raise the price of the TinySA package a little bit. ? This would mean signals up to 1 watt could be safely be used with the TinySA and if a higher level was used it would damage the attenuator but probably save the TinySA.

On 2021-02-04, at 19:16, John Cunliffe W7ZQ <n2nep@...> wrote:

Anyone who would assume the input not to be 50 ohms better not play with RF test equipment and go back?playing in his sand castle…?

Yeah, but at least HP caters to the denizens of the sand castle.

We added a link to the wiki in the box in the hope people would read the specification:?
--

开云体育

not on mine

Here's a better view. :)

73

-Jim
NU0C


On Thu, 4 Feb 2021 20:19:26 +0100
"Carsten Bormann" <cabocabo@...> wrote:

On 2021-02-04, at 19:16, John Cunliffe W7ZQ <n2nep@...> wrote:

Anyone who would assume the input not to be 50 ohms better not play with RF test equipment and go back playing in his sand castle…

Yeah, but at least HP caters to the denizens of the sand castle.
(The max power warning label is the yellow thing on the lower side of the N connector.)

Grü?e, Carsten

One would hope, buy you have to remember the old saying: "Nothing can be made foolproof because fools are so ingenious".

73

-Jim
NU0C

On Thu, 04 Feb 2021 11:47:58 -0800
"Erik Kaashoek" <erik@...> wrote:

We added a link to the wiki in the box in the hope people would read the specification:
--
------------------------------------------
For more info on the tinySA go to

Dear Erik,

There have been a lot of postings on the various nanovna groups, saa2 in particular, about static discharge damage, and adding an esd diode diodes across each input. The?ESD101B102ELE6327XTMA1, seems popular, (and cheap) with a bilateral breakdown voltage of 5.5v.

Would this give any protection from input overload to the Tinysa? It is a tiny device and cannot dissipate much power but the very low capacitance is not so important for a spectrum analyser and maybe a physically bigger one, but with a 5v (or lower) breakdown voltage would be better.

Steve L

Yup
Also "software, a product that is never completed"

On Fri, Feb 5, 2021 at 12:33 AM, Carsten Bormann wrote:

On 2021-02-04, at 17:02, Hugen <hugen@...> wrote:

SA-20020001

Is that the first one from the batch that was made in the year 2002?

SCNR.

Grü?e, Carsten

Sorry, I made a stupid mistake, it's 2021. The serial number should be in SA-YYMMxxxx format, and the correct serial number should be 2101xxxx.??Except for the 137 units that have been sold, the ones produced later will be revised.

On Fri, Feb 5, 2021 at 02:30 AM, Roger Need wrote:

On Thu, Feb 4, 2021 at 08:02 AM, Hugen wrote:

I will continue to pay attention to community feedback and actively improve production.

Hugen,

I have a suggestion for you.? Why not add one of these SMA 20 dB attenuators to the TinySA package?? They are available in China for only a few dollars and would only raise the price of the TinySA package a little bit. ? This would mean signals up to 1 watt could be safely be used with the TinySA and if a higher level was used it would damage the attenuator but probably save the TinySA.

Thanks for your suggestion, it is now available.

It is recommended to add it to the shopping cart and purchase with tinySA to avoid expensive shipping.

On Fri, Feb 5, 2021 at 01:18 AM, Stephen Laurence wrote:

There have been a lot of postings on the various nanovna groups, saa2 in particular, about static discharge damage, and adding an esd diode diodes across each input. The?ESD101B102ELE6327XTMA1, seems popular, (and cheap) with a bilateral breakdown voltage of 5.5v.

Would this give any protection from input overload to the Tinysa? It is a tiny device and cannot dissipate much power but the very low capacitance is not so important for a spectrum analyser and maybe a physically bigger one, but with a 5v (or lower) breakdown voltage would be better.

Steve,? ??
Most of those postings are guessing that its ESD damage in the absence of any other logical explanation.? I worked in EMC for a number of years and we used R&S and Agilent spectrum analyzers costing in the tens of thousands of dollars. The Agilent spectrum analyzers had a fused input in the coaxial connector that limited power.? We purchased transient? ?limiters for all our analyzers.? The commercial limiters from Agilent costed 9x the price of a tinySA.

? ? The thing we learned about limiters is that since a diode is a non-linear device it is subject to acting as a mixer when subjected to pulse signals and creating signals that really aren't there.? Its very embarrassing failing a device because of signals created by your test set-up.? We eventually just annually ordered new 2W 6/10/20dB attenuators that were sacrificial elements against accidental frontend overload.? Much cheaper than replacing the front end attenuator inside the spectrum analyzer.

? I have 3 other USB spectrum analyzers (RFExplorer, Anrinst SSA-TG, and TSA4G1) that all cost much more than the tinySA.? Only the Anrinst device has a power warning label on it and it is pretty much the same as what hugen intends to use.? The fact that hugen is being proactive and adding the warning label says a lot about his commitment to his products and user base.? Add the technical support that Erik has continued to provide and the tinySA is an absolute bargin.

- Herb?

On Fri, Feb 5, 2021 at 08:24 AM, hwalker wrote:

On Fri, Feb 5, 2021 at 01:18 AM, Stephen Laurence wrote:

There have been a lot of postings on the various nanovna groups, saa2 in particular, about static discharge damage, and adding an esd diode diodes across each input. The?ESD101B102ELE6327XTMA1, seems popular, (and cheap) with a bilateral breakdown voltage of 5.5v.

Would this give any protection from input overload to the Tinysa? It is a tiny device and cannot dissipate much power but the very low capacitance is not so important for a spectrum analyser and maybe a physically bigger one, but with a 5v (or lower) breakdown voltage would be better.

Steve,? ??
........
? ? The thing we learned about limiters is that since a diode is a non-linear device it is subject to acting as a mixer when subjected to pulse signals and creating signals that really aren't there.? Its very embarrassing failing a device because of signals created by your test set-up.? We eventually just annually ordered new 2W 6/10/20dB attenuators that were sacrificial elements against accidental frontend overload.? Much cheaper than replacing the front end attenuator inside the spectrum analyzer.

Back to back limiting diodes a the front-end of a receiver or spectrum analyzer can generate mixing products but their level depends on the voltage where the diodes begin to have any significant conduction.? Historically they have been avoided at the front end of a high-performance communication receiver with a low noise floor because they affect the spurious free dynamic range(SFDR) and lower the OIP3.? And the high capacitance of these diodes limited their application to HF/VHF applications.

Today there are many vendors manufacturing ESD protection devices with much higher conduction and clamping voltages and shunt capacitance? that is a? fraction of a pF. They solve the problems described above in many applications.? The one that Steve posted, that is made by Infineon, is one example and they are only a few cents in volume.? I suspect that if these were installed on the TinySa input mixing products would not be seen and therefore? no effect on performance when operating in Spectrum Analyzer and Signal Generator output mode.

There needs to be some sort of ESD protection on the connectors of the TinySA.? Walking around with the supplied antenna attached is just looking for ESD failure.? Years ago I owned a MFJ-259B antenna analyzer and attached a small whip for some measurements.? It was a dry winter day and ESD destroyed the front end.

Roger

On Fri, Feb 5, 2021 at 10:49 AM, Roger Need wrote:

The one that Steve posted, that is made by Infineon, is one example and they are only a few cents in volume.? I suspect that if these were installed on the TinySa input mixing products would not be seen and therefore? no effect on performance when operating in Spectrum Analyzer and Signal Generator output mode.

There needs to be some sort of ESD protection on the connectors of the TinySA.? Walking around with the supplied antenna attached is just looking for ESD failure.? Years ago I owned a MFJ-259B antenna analyzer and attached a small whip for some measurements.? It was a dry winter day and ESD destroyed the front end.

? ?The transient suppressors purchased from Agilent were $400 devices rated from dc -10GHz, yet their documentation still warned (big black letters) that they were subject to generating signals in the presence of complex modulated signals.?

? ?I agree that if all you are measuring is unmodulated carriers or simple AM/FM carriers that mixing components will not likely be an issue.? For those situations, if you want accurate measurements, you just need to characterize the back to back diodes across the range you intend to use them at.? If you only need relative measurements then this is not important.

? The other consideration is that if either diode fails because of an ESD event, you have no indication? and are not protected from future ESD events of a similar nature.? Commercial transient suppressors also have a 3dB series network that fails open if the diode fails.? Your front end is protected and the series open devices gives a visual indication (no signals passed) that a catastrophic event has occurred.

? Grabbing an antenna attached to a receiver without first grounding yourself was a big no-no in our shop and quickly got you put on the remedial training list.

- Herb??

Carsten,

I'm looking to get attenuators to use with various QRP transmitters baur am not sure how much dB of attenuation is needed to use for each of the QRP transmitters with the following watts output:? 0.5 watts, 1 watt. 3 watts, & 5 watts.? ?I will select four 10 watt max input attenuators, one for each of those four transmitters, and have the following attenuations to choose from: 6dB, 10dB, 20dB, 30dB, 40dB.? I will label each one with which transmitter if will be used to provide a safe input to the Tiny SA.

I do have a KAY Elemetrics model 520 step attenuator but have been unable to fins a spec sheet on it to let me know it max input in watts and the min & max attenuation steps in dBs, so I'm hesitant to use it for anything while I am in the dark on it specs.

EuGene, KA5NLY