Ari Padova

  • Aumenta dimensione caratteri
  • Dimensione caratteri predefinita
  • Diminuisci dimensione caratteri
Home Angolo autocostruzione Apparati RX TX RTX FT817nd power (english version)

FT817nd power (english version)

E-mail Stampa PDF
( 2 Voti )

It started one evening, chatting pleasantly at local club when we end up talking about the FT-817 and its power characteristics with battery power.I had previously seen the measurements made by a US ham, and recalled he showed that up to 11V of power the 817 brings out all its 5 watts.

There are also projects of power conditioners specifically designed to power this radio with just as much voltage it needs whilst protecting against overvoltage and reverse polarity which, let's agree, are not very welcome.

This chat has ignited the fire in me of the "I have to go into the bottom of the matter."

The following few days I start with the investigation phase of what is already available.

I notice that almost all of the projects that I find are based on voltage regulators (type 7809 for instance) which certainly lower the output voltage by simply wasting the surplus.

A unique project of which I am aware, instead uses a switching  supply is from KA7OEI, described on his site with lots of refernces to the 817. That project does not only adjust the supply voltage but it also reduces it to 8V when the radio is receiving to save even more power.

But just reading carefully the site in question I noticed small discrepancies with the measurements, definitely approximate, which I had done some time before to see how the radio behaves feeding it with 3S LiPo cells (which have a characteristic voltage of 11.1 V).
I also noticed that everyone is talking about the FT817 while I have a ND and it is known that this version has a modified power amp.And so, I was already thinking about how to make all the necessary measurements to unravel the doubt.

The project

The ultimate goal that I set, beyond the purely academic question, was to get the most from the batteries that I use for portable activities, particularly in the SOTA activations where the weight and the reliability of the equipment are essential.

As mentioned, I decided on 3S LiPo cells for their weight and, thanks to the advice set out in the forum dedicated to the 817 on Yahoo (, I bought the 3.6A/h weighing 200g from germany made for RC models that can be fitted, if desired, inside the battery compartment of the radio. In fact I keep them outside but..
I might have chosen batteries of higher capacity, thus avoiding to bring more than one when doing activations but, since failures happens, having multiple batteries limits the risk of remaining without power supply in case of problems.In any case, the idea of squeezing the most from these without waste and without losing power teased me too much.
I really like the project from KA7OEI but it suffers from a couple of flaws not inconsequential: first of all he uses older components hard to find, which incidentally certainly cannot compete with newer components in terms of efficiency, and the circuit is a buck (that is a down-converter) designed to work with power of over 12V, yet my batteries go well below this (for the 3S cells minimum discharge voltage is 9.9V).

The idea that was flashing in my head was then to design a buck-boost by exploiting the idea of KA7OEI to vary the output voltage from the converter when the radio is receiving in order to obtain the maximum energy savings.
The first thing to do, however, is to decide what should be the power supply voltage during transmission, so all the necessary measurements are required to find the minimum value that allows the radio to go out at full power.


For the measures I used the oscilloscope I have available that reaches up to 70MHz (and is also the only calibrated instrument I own) to measure the voltage across a dummy load with one channel and simultaneously the power supply voltage with another.

I did some random measurements in advance to plan everything as best I could. At this stage I have verified that the power consumption does not change significantly at different supply voltages and settles down to around 300mA in receive and 2A during transmission.
I have noticed however, that the behavior on each band is different and so I decided that I would carry out the measurements on all bands available gradually lowering the voltage by about a tenth of a volt down to the limit tolerated by the radio.

Given the large number of measurement points I proceeded to automate the procedure as much as I could. I wrote a simple program in python that, after setting up the oscilloscope, it performed, in order:

  • commands the radio on the band center

  • Sets the time scale as a function of frequency

  • Tells the radio to transmit

  • Waits for the end of the acquisition by the oscilloscope

  • Stops the transmission

  • Reads the 2 measures of voltage and the frequency of the trigger for verification and then saves all them to a CSV file

  • Repeats from the beginning for all the bands, 4 measurements each

  • At the end ask me to change the power supply voltage and at the push of a button begins again

At the end of the testing, I found myself with 1500 measurements ‚Äč‚Äčof the rms voltage on the dummy load, I had then to apply the well-known formula P=V^2/R and bring everything on a graph. The result is shown in the figure, we can see immediately that for each frequency the power remains almost constant until the supply voltage passes a threshold, different for each band, under which the power decreases linearly. For lower frequencies the voltage threshold is down to almost 8V at which point, the radio simply turns off.

Before starting with the actual design I felt I should evaluate what would be the energy saving thanks to the converter, specifically if the savings achieved with the dual voltage TX / RX justify the greater complexity of the implementation.

As the first hypothesis I considered that the voltage 9.9V for transmission was the right compromise to be able to use a simple (and more efficient) buck rather than a buck/boost. The power loss on all bands is still limited and I find it acceptable.

I finally assumed an efficiency of 95% for the converter and a duty cycle TX/RX of 25% and thus obtaining a saving just over 10% - somewhat disappointing considering that these hypotheses are decidedly optimistic.

It should also be noted that a switching supply would require special attention in the implementation to avoid introducing noise in the receiver, this truly was the aspect of the design that worried me the most.

Finally, if we think back to the saying "what is not there does not break" we should definitely bring an extra battery ... and nothing more!


While I was there I redid the calculations with slightly different assumptions: a 13.8V voltage from the battery and a duty cycle of 10% that I considered representative of portable use but less stringent than a SOTA or other activation.

In this case, you can end up to a saving of 32% but the greatest gain due to dual power supply is only 8%, which in my opinion does not justify the added complication of the setup.

As a result with these assumptions it would be interesting to see the results of using a simple but highly efficient converter .


To summarise I would then confirm that the choice made ‚Äč‚Äčby many users of the 3S LiPo batteries, with no power conditioner used as a portable power source for use on the 817 is best and does not involve significant losses of transmit power even squeezing the cells to their maximum discharge voltage (3.3V per cell).

For use with lead-acid batteries instead (or with any higher voltage supply) it would be worthwhile to introduce a voltage converter, but this is not the route I chose.

73 de IZ3MGE Marco

Tnx to Mike M0SAZ for his support with english translation

Ultimo aggiornamento Domenica 14 Settembre 2014 09:21