开云体育

I would understand that this is not possible due to copyright limitations so as not to make it easy for cloners, but a TECHNICAL DESCRIPTION with simplified block diagram, like the one for TinySA classic on the wiki, would be much appreciated as it would help us to understand how the ULTRA works.

Toni

The charging circuit of tinySA ULTRA is designed for a 3.7V battery and will stop charging when the battery is charged to 4.2V. If you plug in a 3.8V lithium battery, you can only use about 80% of the power of this battery because the charging circuit cannot charge to the full voltage of 4.3V of a 3.8V battery, so you will not get much improvement in battery endurance by replacing the battery.

My question would be can the charging circuit in the TinySA handle the bigger battery?

On Sun, Mar 5, 2023 at 01:29 PM, Святослав Маханьков wrote:

I meant inside tinis ultra.

? ?If you mean AD8317 Logarithmic Detector.?The tinySA ultra does not have a discrete power detector. The power detector is inside the SI4468 IC, after the selectable resolution bandwidth filter.??

? ?The only discrete amplifier used is the?SPF5043Z LNA.

? ?Amplitude calibration uses the SI4468 RSSI output and has a dynamic range of -120dBm.

? Out of the box amplitude accuracy should be +/- 2dBm.

Herb

I meant inside tinis ultra.

Why can't I use the ad8317 logarithmic amplifier IC for calibration?

Martin, thanks for the clarification.

Erik, you specified the threshold very precisely.
6.22mVp-p (-40dBm), actually displayed as -39dBm:

Single spikes are visible and indicates a red AM.


5.09mVp-p (-42dBm), actually displayed as -41dBm:

No spikes are visible and there is no red AM.
It works, thanks for the advice!

With best regards,

I mostly try to use it for Phase Noise measurements of my SHF oscillators.

i2NDT Claudio

On Sun, Mar 5, 2023 at 12:43 AM, Ctoyan Dombarov wrote:

I hope for advice on what can be done to eliminate spikes in a problematic device.

Try to keep the input signal below -40dBm This may solve the problem with the spikes

?
--

The Tiny SA is only specified for frequencies >100 kHz.

I notice the sweep is from 3 kHz to 103 kHz.

I'm surprised you have been able to obtain a trace that looks anything as good as the ones you have, when used at that low a frequency.

The differences in traces is probably due to minor hardware variations and internal noise generation, as it is being used well outside of its specified range.

I think you just got lucky with one device performing better than the other, due to a variation in manufacturing tolerances.

Regards,

Martin

Hello

Erik, you're absolutely right.
I have periodic access to the HP3588A spectrum analyzer, on which I tested my HP3314A function generator.


This 3314A served as a source in previous experiments.

Isidro, the characteristics of the input signal are as follows:
The form is sinusoidal
Amplitude 35.6 mVp-p??????
The frequency sweeps linearly and smoothly from 3 kHz to 103 kHz
The sweep time is 1000 seconds.

I did this to more precisely adjust the frequency correction in the low frequency region.
I didn't expect such different behavior from my two instances of tinySA Basic.
I hope for advice on what can be done to eliminate spikes in a problematic device.

With best regards,

pe1hwo

He is measuring the maxhold of a frequency sweep to try to see the flatness of the level measurement over frequency
--

The correction tables are generic for all same model tinySA and may improve over time. Unless you have created your own correction tables, you should never need to save/restore them over a FW update
--

What are the characteristics of the input signal?
--

Depth error is a warning that with the currently requested output levels there is insufficient headroom to perform the requested AM modulation.
The device is not faulty but you are running into a limitation.
Try to reduce or increase the output level and see if the warning disappears
--

开云体育

Hello,

I don't have a clue what you try to measure or demonstrate.

I can't identify any signal.

Can you please tell us what signal you measure and what you expect?


All the Best,

Isidro

The measurements continued. 1 swiping cycle of source took 1000 seconds.
Accumulated 4 cycles:

Thank you. I seem to have missed this thread.
There was a question on correction. Is it worth using the previously recorded correction factors in the new firmware?