Interesting Spektrum DX3r Experiment

[BrianG]

A little bored and no projects I can work on (or should I say, have no money to work on), I started thinking of the "ideal" lipo solution for the DX3r. But before I can do anything, I need to get some data first. So, I hooked up my home-made adjustable power supply to the DX3r's battery terminals and fired her up.

I found that the radio will power up at 2.6v, but does not transmit. Current draw during this time is ~70mA. Once the voltage gets to 3.1v, current jumps to 127mA and starts transmitting. What I thought was interesting is that the current stays at 127mA all the way up to 6.5v, at which time I did not want to go any higher.

What does this mean? Well, I suspect any voltage above 3.1v does not increase transmitter range. You may say "But as the voltage goes up, the power goes up even though current stays the same, so the range has to be better with higher voltage". The first part is correct, but not so sure about the range assumption. Why? Because I highly doubt there is circuitry in there that is designed to draw X current no matter the voltage. Just doesn't make sense.

There isn't a switching regulator because as voltage goes up, current would go down.

The only logical explanation (to me) is that there is a linear voltage regulator inside set to around 3v. So no matter how high the voltage gets, the extra is being burned off as heat, but since current is only 127mA, that heat is minimal. So, I really think anything above 4 AA cells is just a waste. Kinda makes my 5-cell mod a little pointless. Unfortunately, testing the range is a bit cumbersome since I'd have to have someone hold a vehicle at various distances to test. Maybe someday.

Another interesting thing I found is that the volt meter on the radio isn't accurate. The display read a constant 3.3v when I applied 2.6v up to 3.4v. After that, the display reads 0.2v lower than reality. Hmm, I wonder if that's the v drop of a germanium diode?

Anyway, just thought I'd share my findings...

[Ryu James]

very interesting. great use of scientific principles. this must be what ben franklin and thomas edison would be doing if they were here and into rc. someday i'd like to understand all that stuff better....or just leave it to the experts, like you. interesting findings though.

[suicideneil]

I wander if you could trace the voltage meter back to somewhere, then rewire it so that it takes a reading more directly (obviously not from the batts directly as it would always be on & draining them down slowly)?

[simplechamp]

So this means using single cell lipo power could be possible?

[BrianG]

Originally Posted by suicideneil

I wander if you could trace the voltage meter back to somewhere, then rewire it so that it takes a reading more directly (obviously not from the batts directly as it would always be on & draining them down slowly)?

Not sure if that would be adviseable. Easier (and safer IMO) to just remember that the display will read 0.2v lower than the actual voltage, until it gets down to under 3.5v.

Originally Posted by simplechamp

So this means using single cell lipo power could be possible?

I still haven't experimented with the range vs voltage part yet. It would be nice though as no balancer, regulator, or lvc (use radio warning) would be needed. Also, if there is a linear regulator in there, you have to consider the drop-out voltage.

[brushlessboy16]

Since the voltage is going up, and the current stays the same- watts of output go up? Increasing range?

Because like you said- current would go down with higher voltage only IF they wanted to maintain the same power output.

[BrianG]

With a static load, increasing voltage increases current. The only way for current to stay the same for an increasing voltage was if the load changed, and I don't see that happening. Besides, why exactly 127mA? Why would the circuit just choose that exact current and then adjust the load?

[redshift]

I'd say this is one of the best uses of free time I've seen from anyone lately!

I am very interested, I have a DX5e. I won't use Nimh, because it drops to below 5V pretty quickly. This was a concern for range, so I have been feeding it alkalines to keep the voltage around 6. But now you say there may be no real need for it... that would be GREAT!

The 8 cell (12V) TXs are cake to do lipo with, obviously. But the 4 cell TXs have been a problem haven't they... BTW I thought your 5 cell mod was awesome Brian, even if it didn't end up gaining anything range-wise, it was great for more capacity.

I'll be watching this one keenly.

[bdebde]

I had read somewhere that spektrum will have full range even at minimum voltage, unlike older radios that were effected by voltage.

[redshift]

A few questions, for anyone.

What is the average amount of headroom on components?

20-25%?

Has anyone tried 2S lipo with a TX requiring 6V input?

A full alkaline AA can read over 1.6V new, I've seen as high as 1.7. This equates to 6.8V as a potential high voltage for 4 cell. Taking this into consideration, how risky would an extra 1.5V be? Full charged a 2S is 8.4, only 1.6V over the (I think) obvious headroom given in a 6V application.

And finally, does anyone dare me.... hehe

[BrianG]

Well, I emailed Spektrum and they specifically said 2s lipo, (8.4v peaked) is too much. Yeah, NiMh can get in the 1.5-1.6 range but it is only that high right off the charger and drops pretty quickly to the 1.4v range.

[MetalMan]

Might I approach with the 5v switching BEC idea? (again)

[redshift]

Ok thanks Brian, won't do that. I like 1 cell better, if that will work. Nothing to balance.

MetalMan I know many people do use BECs, I had a BEC fail on me immediately, before I could use it. A failure with a BEC in a ground vehicle only risks other components. But a failure in a TX (specifically an air TX) could mean the loss of both the TX and the plane/heli. Not willing to risk that, I would risk a range loss at low voltage first lol.

[MetalMan]

To each their own I suppose, although I've been using a cheap ~$10 5V UBEC in my React for well over a year now with 2s Lipo. As Brian noted the current draw is extremely low, and thus the chance of failure also should be.

The only downside (IMO) of the UBEC/BEC method with 2s Lipo is that you need to be able to disconnect the battery. I have a second switch that does this for me, although there have been times where that switch was accidentally left on. This has killed one 2s, and nearly another.

[BrianG]

Ideally, any circuitry should be installed after the switch.