Various 5.8 GHz TX measured power outputs

[Big Island Guy]

Well, this is turning out to be interesting.
Using an Immersion power meter (yeah, I know, its not quite like a Bird, but it still gives a reasonably accurate reading) I measured some of the TXs I have around here.

250mW rated Boscam - actual output: 166 mW
500mW rated DJI specimen 1 - actual output 1204 mW
"" """ "" ""specimen 2 - actual output 1077mW
2000 mW rated Boscam - actual output varied radically across the band (confirmed on the spectrum analyzer and by feeling the attenuator which stayed cold on CH8 but rather quickly got warm on CHs 1 - 4 where output was high)
Ch4 = 5645MHz output: 2100mW
Ch5 = 5885MHz " " 493 mW
Ch8 = 5945MHz " " 24 mW

Here is a video of the relative output levels on the spectrum analyzer as I switch through the channels. You can see the output drops as the frequency goes up. It doesn't look that drastic on the analyzer because the display is logarithmic. Tomorrow I'm going to circle back and run all the transmitters through their freq. ranges to see if they also share this unfortunate characteristic.

[setishock]

Concerning the radiated power dropping off as you switch up through the channels, the higher the frequency the more power it takes to generate that signal. So if the power input remains constant and the frequency goes up the amount of radiated power would decrease.
I would venture the VCO is not the problem. It seems like the RF amp is being fed a constant current to keep the maximum output power capped.
If the RF amp was fed by a regulator with a feedback sensor loop, as the VCO frequency rises, the regulator would open up giving the RF amp more power to generate the higher frequency signal.

The output power dropping off as shown by your test results, is indicative of a fixed power system. I know it's illegal to do this but maybe look in to strapping out the finals? If you try that you need to have a dummy load on the antenna. You don't how much power the transmitter can actually produce. And it's microwave. You could get a nasty RF burn.

To be clear when I say power I mean the battery or power supply voltage being fed to the transmitter.

[clouseau]

Was the power source the same for all the transmitters tested?

I've noticed that some transmitters output wattage was varied in relation to the input voltage.(power supply)

Next test you may want to try is with a bench top variable power supply.

Measure the Transmitters output at the lowest required input voltage, then again at the highest and see if that makes a difference in your numbers.

The better transmitters should not change output wattage throughout their voltage input range.

[truraza]

I would like to know how the fatshark stuff works...
DJI seems to be consistently "under-rated" based on your two samples.

[Big Island Guy]

Originally Posted by clouseau

Was the power source the same for all the transmitters tested?

I've noticed that some transmitters output wattage was varied in relation to the input voltage.(power supply)

Next test you may want to try is with a bench top variable power supply.

Measure the Transmitters output at the lowest required input voltage, then again at the highest and see if that makes a difference in your numbers.

The better transmitters should not change output wattage throughout their voltage input range.

Same B&K Precision bench top supply, at same voltage (12.4) for all four TXs, which is the max allowed for the Boscam 250mW, but not the Boscam Thunderbolt (the 2W one) or the DJIs that can run up to 6s. I double checked on the Thunderbolt to make sure the PS's current limiting wasn't kicking in. It wasn't, although the TX does draw less at the top end of it range, where it's putting out less. It would be interesting to run through their voltage ranges though. Also, tonight, I'm going to see if the other TXs are consistent when I change frequencies, something I didn't get a chance to do last night as it was getting pretty late.

[CircleH]

I like reading the conversations you guys have about radio tech.. I have very little understanding of it other than it is all part of the EMS. Maybe one day, after having listened to and read all that you guys contribute... I will have that "Eureka!!" moment....and it will all fall into place. know of any good "primers"?

thanks.guys.

Rich.

[z_man_95]

Originally Posted by CircleH

I like reading the conversations you guys have about radio tech.. I have very little understanding of it other than it is all part of the EMS. Maybe one day, after having listened to and read all that you guys contribute... I will have that "Eureka!!" moment....and it will all fall into place. know of any good "primers"?

thanks.guys.

Rich.

+1

[Big Island Guy]

Some further notes:
First a correction - the Boscam that I said was rated for 250 mW is actually advertised to put out 200mW.
With further testing tonight, I noticed the power output dropped a bit as they warm up, so I let them do so.
The Boscam 200mW was fairly steady across its band. It is designed for 12V so there's no voltage variation testing to be done. With 12.4 V input after warm up it showed:
5645 = 128 mW
5885 = 109 mW
5945 = 100 mW

The DJIs were consistent at right around 1 Watt across the 8 channels and rated supply voltages.

The Boscam Thunderbolt was also stable across it's working voltage range, but a complete dud on the higher frequencies as noted above.

I also tested a little Skylark TX that's supposed to put out 1W, which it does when cold, but it rapidly gets very hot (as Skylark also notes) and when it does, the output drops by about a half, to the 500 - 600 mW range across all frequencies. It is only made for 5V, so no voltage range testing done.

Another thing I noticed, it that in all of these except the Thunderbolt, there are discrepancies in what frequency the documentation says a given switch position should set them to and the reality. For example, the Skylark TX claims 5945MHz is switch position #16, but in reality it's at position #14. I was wondering why sometimes what my goggles should have been seeing wasn't where it was supposed to be. Ideally, if one is running two 5.8TXs, it would be good to space their outputs as far apart as possible, but it's kind of hard when the documentation is wrong.

I'm learning to take all the specifications with a grain of salt and measure for myself. Next, I'll test my 1.3GHz VTXs.

[CircleH]

So.. with your observations, would you consider the DJI stuff to be the "Winner" ?

[Big Island Guy]

Originally Posted by CircleH

So.. with your observations, would you consider the DJI stuff to be the "Winner" ?

Definitely, at least going by the two samples I tested. Not only are they the only ones that put out what they're advertised to (actually double what's advertised!) across the band, but they're very nicely cased too, with decent cooling and very easy to select channels, with one button instead of three or four tiny little fidget switches. They also are voltage versatile, good for 3S - 6S battery stacks. I wish they'd offer them on other bands. About the only drawback, besides DJI's usual premium price, is that like most of the others, the frequencies specified in the manual as corresponding to the channel switch settings do not all seem to line up with where they're actually at frequency wise so you'll have to hunt around a bit, as I did, if you're not using DJIs receiver.

[CircleH]

Maybe thats why they have that "premium price" Im not sure I like that they are so expensive and most electronics today (i think anyway) are easy to manufacture... and should be a lot cheaper than what they actually cost the end user..

funny.. The very first gear I looked at was the DJI stuff.

Thanks for posting this.. Good info to know. I read an article about lawmate being an unreliable product.. Was going to go with them but now I find myself with second thoughts.

[Big Island Guy]

From what I understand, Lawmate was a good brand, coming highly recommended by various FPV folks, but apparently something happened because recently I saw a notice on readymaderc's listing that they were dropping Lawmate because of quality control issues coupled with poor customer service. Now that notice is gone and they're still carrying Lawmate, so go figure eh? Naturally, with my luck I only found out about this after ordering a Lawmate 1.3GHz TX/RX combo from Wireless Hobby because Dragon Link was recommending those. Oh well, maybe I'll get lucky with it. I'm going to measure it's output on a few minutes. Just in case, I ordered another 1.3 from another source. That one supposedly puts out 2.5 W. We'll see soon as I have right here, ready for measurement..

[jamesb72]

Those numbers look a bit odd, are you sure the power meter is accurate? Hard to understand how changing channel would alter vtx output by 2 orders of magnitude?

[Big Island Guy]

The 2500mW rated 1.280/1.258 GHz TX from ReadymadeRC tests out at 2300mW - essentially as advertised. An interesting illustration of why one needs to be careful running 2.4GHz controls with these VTXs on board can be seen in the these two pictures of the spectrum analyzer. The first is what it's seeing on the 2.4GHz band, with a little 2dbi rubber duck and the 1.3 TX off. You can see to main concentrations of energy from the two internet access points in the house. The second picture is what we see with the 1.3GHz TX turned on, about 2 feet away - actually a little more separation than a typical MR would allow for. Notice the new band, right around 2.16 GHz, the 2nd harmonic of the TX transmitting at 1.258GHz.

2.4GHz ambient, 1.3GHz TX off.

2.4GHz ambient with 1.3 GHz TX on.

[Big Island Guy]

Originally Posted by jamesb72

Those numbers look a bit odd, are you sure the power meter is accurate? Hard to understand how changing channel would alter vtx output by 2 orders of magnitude?

That crossed my mind and I might have thought so but for four things 1)It's not consistently low on other TXs at the same frequencies. If there was a consistent bias, I'd suspect this too. Secondly, as shown in the video the spectrum analyzer reflects the drop off. Thirdly, the current pulled from the power supply falls off proportionally and lastly, as I also mentioned, the attenuator gets warm quickly when it's putting out rated power and stays stone cold at the other end of the band. If it walks like a duck and it quacks like a duck, then....

There are a lot of reasons a TXs output can fall off drastically when the frequency is changed. Modern TXs can automatically tune themselves, but if something's not right, all bets are off. I'll leave it to you to find out for yourself some of the ways this can happen.

[clouseau]

Originally Posted by Big Island Guy

The 2500mW rated 1.280/1.258 GHz TX from ReadymadeRC tests out at 2300mW - essentially as advertised. An interesting illustration of why one needs to be careful running 2.4GHz controls with these VTXs on board can be seen in the these two pictures of the spectrum analyzer. The first is what it's seeing on the 2.4GHz band, with a little 2dbi rubber duck and the 1.3 TX off. You can see to main concentrations of energy from the two internet access points in the house. The second picture is what we see with the 1.3GHz TX turned on, about 2 feet away - actually a little more separation than a typical MR would allow for. Notice the new band, right around 2.16 GHz, the 2nd harmonic of the TX transmitting at 1.258GHz.

This is neat being able to see the 2nd harmonic, which is something many guys don't understand.

This is why it's said not to use a VTX with a lower frequency than your radio control frequency. It's also why 5.8gHz is so popular, since the majority of all radio control systems operate at 2.4gHz.

There are many claims out there that you can use 1.2 - 1.3gHz VTX with a 2.4gHz radio receiver with a low pass filter, however if you don't have the equipment as Mr. Big does here... the question you have to ask yourself is if it's really worth loosing your multirotor not knowing for sure if your low pass filter is working or not.

Very cool work Big Island Guy

[clouseau]

Originally Posted by Big Island Guy

That crossed my mind and I might have thought so but for four things 1)It's not consistently low on other TXs at the same frequencies. If there was a consistent bias, I'd suspect this too. Secondly, as shown in the video the spectrum analyzer reflects the drop off. Thirdly, the current pulled from the power supply falls off proportionally and lastly, as I also mentioned, the attenuator gets warm quickly when it's putting out rated power and stays stone cold at the other end of the band. If it walks like a duck and it quacks like a duck, then....

There are a lot of reasons a TXs output can fall off drastically when the frequency is changed. Modern TXs can automatically tune themselves, but if something's not right, all bets are off. I'll leave it to you to find out for yourself some of the ways this can happen.

Just something to consider...
If you are using a tuned antenna, the lower wattage numbers and heating up you are seeing as you change frequency band is moving further away than what the antenna is tuned to. (VSWR match)

If you have a VTX operating on a specific frequency, such as 1.3gHz. In order to transmit for best output power it's best to keep the VTX as cool as possible and also use the frequency with what the antennae is tuned.

If you use just a rubber duck style antenna it will still work no doubt, however it will heat up the VTX more, which will in turn also decreases the power output. (wattage)

So using a tuned antenna and operating your VTX at the frequency the antenna is tuned to, should help keep you VTX cooler, while not cutting your power output as drastically.

Lawmate mods
http://beta.ivc.no/wiki/index.php/Lawmate_Mods

[setishock]

You'd have to backtrack the length of exposed wire on the cloverleaf antennas to be able to find the frequency of the band it's on. Mass produced, off the shelf antennas will have some "slop" in them. The cloverleaf's I have are simply advertised to be on the 5.8ghz band. As to finding the exact channel would require measuring each leaf and averaging that then run the numbers or consult a known good chart.

If dealers had to supply a directly tuned antenna for each channel their inventory would be fairly large and the cost per unit would go up. Besides being a nightmare for the consumer having to figure out what channel they will be operating on and making sure they order the right antenna. So in light of that it appears suppliers are simply marketing a broadband type of antenna. If you want to lock your antenna on a set frequency, you wind up making your own and exercise a tighter tolerance.

Unfortunately not every one has a micrometer to measure out the length of wire to match the frequency and/or has the soldering skills to make a decent solder joint. So they end up having to use off the shelf, exact channel unknown, commercial antennas.

You can hop through the channels on your RX/TX until you get a decent quality image and that would give you an idea of the channel the antenna is tuned to. But if the antenna construction is not on the dime, it still might be off to one side of the channel.

[ErichF]

I know from practical experience that the DJI Data Link radios are more powerful than they suggest. I have flown the 900 data link out as far as I can still have enough battery to return home and never had less than four bars on the RSSI. The 2.4 datalink I have flown out to over 1KM and still had three bars RSSI. One day I will do a no-chit range test of these data links on a forest road my company uses to range test our other UAVs. It's a perfect straight line that's 16KM long.

[Big Island Guy]

Originally Posted by ErichF

I know from practical experience that the DJI Data Link radios are more powerful than they suggest. I have flown the 900 data link out as far as I can still have enough battery to return home and never had less than four bars on the RSSI. The 2.4 datalink I have flown out to over 1KM and still had three bars RSSI. One day I will do a no-chit range test of these data links on a forest road my company uses to range test our other UAVs. It's a perfect straight line that's 16KM long.

Erich, I'll test my 900 data link tonight to see what my stuff shows it putting out. I wish I had an empty road like that to test with!