开云体育

You can use NanoVNASharp MOD V3:

It works on Windows XP and has most of needed functionality, include capture screenshot button and TDR :)

In general:

The death of an IC or damage when subjected to voltages outside its maximum ratings is not *guaranteed*. That explains why the VNAnano does not die from Vbat of 4.01V, nor 4.1V or even 4.5V. But it's not guaranteed so stating "my VNAnano survived 4.2V at Vbat" means nothing more than that *that* VNAnano survived it and at *that* time, *that* temperature etc.

As can be seen in the data sheet:

Re. Vbat:

- Vbat is *never* allowed to exceed Vss by 4V (Absolute Maximum Rating). So the Vbat pin's maximum in the circuit is 4V. That's what D2's cathode is connected to.
- The full Vbat is used to supply backup power for 5 registers and for the RTC. Without Vbat connected (without D2 inserted) these registers and the RTC do not receive power when the VNAnano is powered down. Any data that survives a power down condition is *not* stored in any of these registers.
- Only for the purpose of measuring battery voltage, Vbat is branched off and divided by two inside the IC in a 50% voltage divider (2 x 50 kOhm in series). This is done to make sure that the ADC input that is used for measuring Vbat remains within range. BTW: as a result of this, if the internal divider remains connected during power down (not sure), the battery will be supplying about 40 uA max. through the voltage divider and the ADC input pin would remain at 2V. Not sure if that's OK....
- A diode (both a normal Si diode and a Schottky) between battery and Vbat will normally keep Vbat below its 4V max. Since these two different types of diode have different drop voltages (about 0.7V and 0.2V respectively, depending on construction, model and current, the difference between battery voltage and Vbat must be entered in the software to get the correct battery voltage indication.

Re. protection against overcharging:

Using a zener diode shorting the battery is a bad idea, mainly for the following reasons:
- A zener (especially real zeners as used at this low voltage) does not have one voltage point where it suddenly starts conducting but rather, a trajectory where it goes from (almost) no conduction, via slight conduction to stronger conduction. This is not linear: around the zener voltage, the current vs. voltage curve steepens strongly. As a result, a 3.9V zener diode starts conducting at say 3.6V. The "zener voltage" is specified at a certain conduction (a number of mA). If you want to use the "3.9V" zener at a current that's significantly different from the current that is used in its voltage spec, it's not a "3.9V" zener for you.
- Like all components, zeners have a tolerance in their zener voltage. The 1N4730A thas was suggested elsewhere has a 5% tolerance, which is pretty good. It means however that the actual zener voltage (the "knee") may lie between 3.7V and 4.1V for this 3.9V zener (@ 64 mA). Not very useful for our purpose.
- The 1N4730 is a 1W zener, meaning that its maximum allowable current would be 250mA (keeping the package cool). I don't know what the charging current in the VNAnano is that would be shorted by the zener but it could well exceed 250mA.

It makes no sense showing the battery voltage in mV. My personal preference would be showing Volts (so not mV) and one decimal.

I'm sure I forgot some points....

Raymond

What diode are you using? Thanks...
73

--
On the banks of the Piscataqua
Rich NE1EE

I'll try again...it would not install on my Android 6x device, but I'll revisit.

~R~

--
On the banks of the Piscataqua
Rich NE1EE

For those intrepid souls who want to push the limits of the older Android devices, you can download the NanoVNA Android app latest release from the developer's github webpage:
?? Latest release is from 3 days ago:?

First make sure developer options are enabled:?

Enable developer options and debugging
- Open the Settings app.
- (Only on Android 8.0 or higher) Select System.
- Scroll to the bottom and select About phone.
- Scroll to the bottom and tap Build number 7 times.
- Return to the previous screen to find Developer options near the bottom.
Then, you need to enable side-loading through developer options:

?

If? v1.4 doesn't work on your Android version, try one of the earlier releases.
...Larry

Agreed that this is sensible...
1) I seem to recall that VBAT can go as low as 1.5V and still power the RTC et al. Not verified.
2) We are discussing this only for battery voltage indication, so an offset of say 0.6V will drop the low end of the battery from, say 3.3V? (not sure when the IP5303 quits) to 3.3-0.6 = 2.7V, which, if I recall the 1.5V value correctly, will still power the RTC if we wind up caring.

So VBAT will range from 4.2-0.6 (nominally) = 3.6V to 2.7V(?) before the IP5303 quits. If the software assumes a 500mV drop, and I wind up with 700mV, will I really care? Well, LI batts have a fairly flat discharge curve, so that 200mV difference might count. If I can enter this into the device, I am good to go. All I have to do is measure the diode drop accurately.

I appreciate all the advice. I have the battery icon, but no charge indication, so I am looking at doing something.

~R~

--
On the banks of the Piscataqua
Rich NE1EE

The real reason the voltage is limited to 4V is due to the ESD protection diodes in the device. There is a diode between all inputs and VDD, so if a voltage greater than a diode drop above VDD is applied, the diode conducts and a destructive current from the battery could flow. Also, since the battery has a very low internal resistance, the device’s VDD could even rise above the nominal 3.3V since a regulator does not clamp the output voltage to its design value (3.3V).

Not wanting to flame...but I seem to recall seeing a comment about VBAT seeing VDD+0.6V at startup under some conditions...I'll have to see if I can find it...that seems to imply that VBAT can see 4.2V, but as I am not quoting, this is just by way of chatting. I may have misunderstood.

~R~

--
On the banks of the Piscataqua
Rich NE1EE

Visual C ++ is required to render the graphics, so if you can, there is hope.
I tried to install on windows 7-32 bit machine that lacked the SP1 service pack and couldn't install it.
I tried Python 3.7.4 on the same machine, but it didn't work because of the missing Ms Visual C ++.

Gyula HA3HZ

Why are my two still operating OK at a VBat of a little over 4.195 V after the diode then? I do not see that it has any detrimental effects.

Give me a timeline on when the chip will fail if the Max specs are so demanding. I will let it operate 24/7 for a test, just let me know what will satisfy you. I'll bet neither one fails but I will sacrifice one for science.

I will report back if they ever fail. :-)

Folks,
For the android App you need to be running at least android 6.? I tried to install it on an old samsung galaxy note and, when I searched the playstore, it was not listed.? Checking again with my google pixel 2, the app IS listed.? The difference is the Samsung galaxy note is running android 4.3 and the google pixel 2 is running version 10.
I searched for nanovna to find it.
And, on further looking...? the play store identifies it as the "NanoVNA WebApp"!? ?
Cool!? Learned a new thing again today.??
Thanks and I apologize for the confusion (and being confused).??
Mike de N5BGZ

Oristo, according to STM32F072 datasheet, the safe voltage for VBAT is 3.6 V and it should not exceed 4.0V otherwise it leads to permanent damage.

Li-Ion battery has max voltage 4.2 V. Today I tested it for a new battery percent calculation function for battery indicator and found that max voltage is about 4.172 V, so it very close to 4.2 V specification for Li-Ion.

4.2 - 3.6 = 0.6 V

So, the diode D2 should have at least 0.6 V voltage drop for safe STM32F072 operating conditions.

May be your diode has smaller voltage drop and it leads to "dead" effect, what do you think?

The safe operating voltage for VBAT is 3.6 V, so this is why there is needs diode with voltage drop

According to datasheet VBAT pin has absolute maximimum rating 4.0 V MAX. This limit already takes into account internal dividers and other circuit. If you exceed it it may damage device.

Datasheet says: Stresses above the absolute maximum ratings listed in Table 21: Voltage characteristics, Table 22: Current characteristics and Table 23: Thermal characteristics may cause permanent damage to the device. These are stress ratings only and functional operation of the device at these conditions is not implied. Exposure to maximum rating conditions for extended periods may affect device reliability.

The maximum voltage specified in this table for VBAT pin is 4.0 V MAX.

If you exceed this limit, it may lead to chip damage or quick degradation. It may live for some time with such out of spec condition, but anyway it leads to degradation and unexpected failure at any time. It may happens day later or a week, but datasheet claims that such voltage leads to permanent damage to the device. And there is no reason to not believe that.

I think this is an interesting design discussion, because it can have a long term impact on what we do...
I have not had time to research further, but the point about what it powers is relevant
In fact, it power several items, including the RTC, clock crystal, and backup registers. I don't see the NanoVNA using those yet. Not sure about that, though. If VBAT is not powered, as on my device, then they are not in use.
The voltage divider comment doesn't imply that ALL circuits downstream of the divider are supplied with VBAT/2. They /might/ be, but it may also be true that VBAT/2 is sent to the ADC, and VBAT goes straight to RTC, et al.

Is it possible on this device to lower the battery voltage, or is that set by the battery pack?
The power going to VBAT only powers the RTC et al. Battery power for the kit goes through the IP5303. So it seems that we are free to discuss this indepent of device power.

I don't have a suggestion...but it seems that there is much better info on the table now that can lead to better decisions. One of the obvious ones is just to install a diode with as small a drop as possible, and make the correction available to users. I lean toward a 3.8V <5% Zener or there abouts, but I understand that is more complicated.

Also, aside from gaining a better understanding of this device and its design, I keep in mind that this is really just a rough indicator of battery state...not sure that we need much more than estimates of .25 ,5 .75 1.0 Full charge.

~R~

--
On the banks of the Piscataqua
Rich NE1EE

Ah the mysteries of electronics!? I was bemoaning the demise of my unit
while sitting at the desk.? Then I saw the blue light come on.
? I thought I'd turn it on and now it's working normally

It's Halloweeeeeeeeeen!? Buhahaha

That chip is seemingly related to those in small LED flashlights,
where various switch press sequences select on vs off vs blink,
but nanoVNA lacks pushbutton access.

My nanoVNA played dead immediately after D2 installation,
but recovered when connected to USB.

Mike,
Please have a look at the 4th VBAT match on this webpage:

4V? is the max you can stick on VBAT before the silicon is overly stressed and the magik smoke is released.

...Larry

I'm talking about the voltage the Vbat sense pin "sees" - it goes into a voltage divider for voltage monitoring. Yes the backup battery powers things but at that point it is what it is in the Nano board design. If the battery supply system is designed badly it stands alone as a bad design. But at the Vbat SENSE pin it (IMHO) isn't an issue because of the voltage divider. The diode issue is only for the Vbat sense - for the ADC circuit, as we all know if the diode is missing the backup voltages still are working. This is ONLY for the Vbat input pin for sensing. And as I've noted both my units have been operating for weeks with the schottky diode and they aren't a pile of ashes. :-)

Hi Elia,
The output of the NanoVNA is a square wave consisting of a fundamental below 300MHz and its harmonics above.You would need to take all that into account in addition to the discrete freq steps and power levels.
What I want to test is a broadband noise source instead of using Ch0 as a tracking gen but monitor with CH1.
Erik and Kurt might be better suited to answer your question, though.
...Larry

The Vbat does not "power" anything it just goes into a divider for monitoring

Actually, it DOES power the RTC in addition to being monitored.? What would Vbat-Vss give you? ...? since Vss is Gnd and at 0V and Vbat-Vss is 4V max as specified, you need the diode (or two) or zener so you can measure a 4.2V battery.
What I don't think is necessary is 3 decimal points for the voltage. 1 decimal is sufficient IMO.
...Larry

On Thursday, October 31, 2019, 9:31:29 a.m. GMT-4, Mike_nano <biounit.mike@...> wrote:

I know what it says elsewhere in the datasheet for maximums, but I think it's not that simple for Vbat. Snippet from datasheet -

3.10.3 VBAT battery voltage monitoring

This embedded hardware feature allows the application to measure the VBAT battery voltage using the internal ADC channel ADC_IN18. As the VBAT voltage may be higher than VDDA, and thus outside the ADC input range, the VBAT pin is internally connected to a bridge divider by 2. As a consequence, the converted digital value is half the VBAT voltage.

I think it's not going to hurt the chip because of this -

the VBAT pin is internally connected to a bridge divider by 2. As a consequence, the converted digital value is half the VBAT voltage.

The Vbat does not "power" anything it just goes into a divider for monitoring. I think it's just measuring half the external Vbat and then multiplying it back to the actual value, so it can handle higher voltages (makes sense). In any case, I've had my diode (the SD103AWS part) in two units unit for several weeks and had the things on for hours at a time and it still works the same. I do not think it's a problem. :-)