Author Topic: Volts to dBm - sanity check?  (Read 5663 times)

0 Members and 1 Guest are viewing this topic.

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Volts to dBm - sanity check?
« on: April 03, 2019, 05:17:38 pm »
I'm comparing notes with a friend in another part of the country, where each of us is running tests feeding signals to an SDR from a signal generator.

His HP8640B signal generator apparently has output calibrated in dBm, while my DDS Signal Generator/Counter (brand name Koolertron) seems to be set up for output level to be specified only in volts.

I assume I can convert volts to dBm first by calculating power in watts:

P(w) = V^2 / R

where R is the SDR's 50-ohm load

Then to convert the power from watts to dBm:

P(dBm) = 10 * log10(1000 * P(w))

So for our 50-ohm system, the simplified equation would be:

P(dBm) = 10 * log10(20 * V^2)

And plugging this into my TI-89, I get a few values:

Volts                  dBm

1                       13.0103
0.1                    -6.9897
0.01                  -26.9897
0.001                -46.9897
0.0001              -66.9897
0.00001            -86.9897
0.000001          -106.99

So the value is 20 dBm less every time the decimal of the voltage moves over one place.

Am I doing this right? When my friend talks about his SDR detecting a -147 dBm signal, that would be equivalent to 0.01 microvolt from my signal generator?
 

Offline awallin

  • Frequent Contributor
  • **
  • Posts: 694
Re: Volts to dBm - sanity check?
« Reply #1 on: April 03, 2019, 05:38:29 pm »
try this chart:
http://wera.cen.uni-hamburg.de/DBM.shtml

your "volt" corresponds to "volt rms" in the chart..

not sure about our notation in the dBm formula...
 
The following users thanked this post: elroy

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Re: Volts to dBm - sanity check?
« Reply #2 on: April 03, 2019, 06:31:02 pm »
Thanks for bringing up the distinction between RMS, peak and peak-to-peak voltage.

I see from my signal generator's manual that its voltage output is stated as peak-to-peak. The minimum it is capable of at any frequency is 2 mVpp, which from your chart would be -50 dBm. So it sounds as though if my friend and I are comparing SDR sensitivity at levels in the -100 to -140 dBm range, I'm going to have to insert some attenuators of reliable value.

Incidentally, since my signal generator's manual states that its output impedance is 50 ohms (+/- 10%), I assume I can just use a coax cable to attach it to the 50-ohm antenna port on my SDR and not worry about terminators, as would need to be used with, say, an oscilloscope?

 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #3 on: April 03, 2019, 06:35:18 pm »
First of all you're need to make sure that your receiver has 50 Ohm impedance at specified frequency.
Very often it is not 50 Ohm. You can check it with vector antenna anlyzer.

The second, there are different volts: Vrms, Vpk, Vpk-pk.

So, if your receiver input has Z = 50 + j0 [Ohm], and you're talking about Vrms, then your calculations are valid.

For the reference, with Z = 50 + j0 Ohm:

0 dBm = 0.001 W = 0.2236 Vrms = 0.3162 Vpk = 0.6325 Vpk-pk = 106.9897 dBμV PD = 113.0103 dBμV EMF = S9 + 73 dB HF = S9 + 93 VHF


If your generator is marked for peak-to-peak at 50 Ohm load, then power can be calculated with

P = (Vpk-pk/2)^2 / (2 * R)

For Vpk (peak amplitude):

P = Vpk^2 / (2 * R)

Please note that Chinese DDS such as FY6600 shows invalid voltage. As I know it shows Vpk for open output (with no load at all). It means that you don't know what voltage will be with the load, you're needs to measure it. But if you know output impedance of your DDS is 50 Ohm, then you can just divide the value which is shown by two.

As I know FY6600 has 50 Ohm output, but you're need to check it. In order check, do the following:

1) Select 50 Hz 1 Vpk (1 Vpk = 2 Vpk-pk) sine on your generator, but do not connect anything on the output, leave output connector open

2) Measure RMS voltage on the output connector with TrueRMS multimeter.

3) Connect 50 Ohm resistor on the output connector and measure RMS voltage with TrueRMS multimeter.

If your gen has 50 Ohm output, the value measured on the step 2 will be exact 2 times higher that the value measured on the step 3.

If your multimeter shows 0.7071 Vrms on the step 3, then your generator shows correct voltage.

If your multimeter shows 0.3536 Vrms on the step 3, then your generator shows incorrect voltage (this is Chinese volts). It shows voltage for open output. So, you're need to measure actual value with connected load. If your gen has 50 Ohm output, you can just setup twice higher value on the gen in order to get required voltage.
« Last Edit: April 03, 2019, 10:05:06 pm by radiolistener »
 
The following users thanked this post: elroy

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #4 on: April 03, 2019, 07:21:11 pm »
Incidentally, since my signal generator's manual states that its output impedance is 50 ohms (+/- 10%), I assume I can just use a coax cable to attach it to the 50-ohm antenna port on my SDR and not worry about terminators, as would need to be used with, say, an oscilloscope?

if your SDR has 50 Ohm input, it is true, there is no need matching circuit. But very often SDR input is just marked 50 Ohm, because it is about that value in average for all frequency band. But it's actual impedance may very very significantly depend on the frequency.

You're need to measure your SDR input impedance at specified frequency, I'm sure it will be at least slightly different from 50 Ohm even for very precise SDR.

For example RTLSDRv3 doesn't have 50 Ohm on the input. It's impedance vary in wide range depends on the frequency.
« Last Edit: April 03, 2019, 07:29:23 pm by radiolistener »
 

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Re: Volts to dBm - sanity check?
« Reply #5 on: April 03, 2019, 07:25:37 pm »
First of all you're need to make sure that your receiver has 50 Ohm impedance at specified frequency. ...

Great information, thanks for all the detail. Sounds like I have a few things to run down.

Although there are minor differences in panel controls, for all intents and purposes my CJDS66 Series signal generator is close to identical-looking to the FY6600, so I'm going to assume that all of your caveats (and recommended checks) apply.
 

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Re: Volts to dBm - sanity check?
« Reply #6 on: April 03, 2019, 07:27:48 pm »
You're need to measure your SDR input impedance at specified frequency, I'm sure it will be at least slightly different from 50 Ohm even for very precise SDR.

Good to know -- time to break out the miniVNA.

 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #7 on: April 03, 2019, 07:30:04 pm »
What SDR you're talking about?

Wait, I will calibrate my vector analyzer and will show you how input impedance of RTLSDRv3 vary depends on the frequency.  ;)
 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #8 on: April 03, 2019, 07:35:49 pm »
my friend talks about his SDR detecting a -147 dBm signal, that would be equivalent to 0.01 microvolt from my signal generator?

Receiver sensitivity depends on the noise floor. Noise floor depends on the bandwidth. So, ask your friend - what bandwidth he was used in order to detect 0.01 uV signal? For 500 Hz bandwidth (CW) it is usual value.
« Last Edit: April 03, 2019, 07:37:55 pm by radiolistener »
 

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Re: Volts to dBm - sanity check?
« Reply #9 on: April 03, 2019, 07:46:58 pm »
What SDR you're talking about?

Wait, I will calibrate my vector analyzer and will show you how input impedance of RTLSDRv3 vary depends on the frequency.  ;)

In this case we're comparing Airspy HF+ units. I have one that's been exposed to a certain level of nearby RF from a 100w transceiver in recent weeks. I think it should be ok, because it was unpowered and disconnected from its antenna when the transceiver was being used. However, it seemed somewhat less sensitive than I'd remembered when I was using it this week, so I was looking for ways to test it. My friend who also owns an Airspy HF+ offered to do a long-distance benchmark if we could devise comparable inputs.

Your point about the bandwidth is well-taken, and I'll make sure that we look at that.

 

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Re: Volts to dBm - sanity check?
« Reply #10 on: April 03, 2019, 08:12:12 pm »
Please note that Chinese DDS such as FY6600 shows invalid voltage.

Incidentally, my friend echoed the observation that the Chinese DDS has severe shortcomings. I knew from the outset that it wasn't a real quality instrument, but I wanted to see if I could get some use out of it and hoped it wasn't a complete loss.

If I want to replace it with a better signal generator, are there any makes/models you would recommend? Knowing that I'd need to spend more, but cannot afford, say, thousands of dollars.

 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #11 on: April 03, 2019, 09:58:35 pm »
As I know, regarding to amplitude, FY6600 works pretty good. But you're need to take into account that it shows Hi-Z voltage for open output (with no load) and when you connect 50 Ohm load it will be divided by two. But this is true if your gen has 50 Ohm output. Just check it and if it is 50 Ohm, then just setup twice higher voltage than you needed and it will works.

There is some noise on FY6600 output from switching power supply, but it can be fixed with replacement switching power supply with linear regulated one.

The main problem with FY6600 is a jitter due to FPGA code implementation. It leads to parasitic spurious components in the signal.

If I want to replace it with a better signal generator, are there any makes/models you would recommend?

Siglent SDG2042 nice gen and it shows voltage as expected, you can select Hi-Z or 50 Ohm indication. But it is more expensive.

If you wan't something more cheap, I think it's better to buy some used but brand name equipment
« Last Edit: April 03, 2019, 10:08:54 pm by radiolistener »
 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #12 on: April 04, 2019, 12:24:18 am »
Here is input impedance for genuine RTLSDR Blog v3 (bought in official store). PCB version 2018  9  13.
https://www.rtl-sdr.com/buy-rtl-sdr-dvb-t-dongles/

The dongle is powered and switched into direct sampling mode. LO tuned at 5 MHz.

As you can see it is not 50 Ohm and significantly varying depends on the frequency.

Impedance for 1 - 81 MHz:


Impedance for 0.5 - 17 MHz:


VSWR for 1 - 81 MHz:


VSWR for 0.5 - 17 MHz:

« Last Edit: April 04, 2019, 12:28:50 am by radiolistener »
 

Offline elroyTopic starter

  • Regular Contributor
  • *
  • Posts: 90
  • Country: us
Re: Volts to dBm - sanity check?
« Reply #13 on: April 04, 2019, 12:31:15 am »
Looking at the plots reminds me of the old saying: There are two times a day when a broken clock has the correct time ...

 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 4135
  • Country: 00
Re: Volts to dBm - sanity check?
« Reply #14 on: April 04, 2019, 12:44:12 am »
Looking at the plots reminds me of the old saying: There are two times a day when a broken clock has the correct time ...

something like that ;D 
But if you look at SWR plot, it is not so bad actually :)
And it works with pretty good sensitivity for it's price.
« Last Edit: April 04, 2019, 12:46:18 am by radiolistener »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf