RC Groups

RC Groups
    NQRC / PLDaniels Electronics
        Build Log PLD-BESJ5AR: 5A linear regulator with alarm, switchjack, voltage readout, decoupling

#1 PLD Nov 08, 2011 05:01 AM

PLD-BESJ5AR: 5A linear regulator with alarm, switchjack, voltage readout, decoupling
 
1 Attachment(s)
This one has been sitting in my "might build" archive for a loooooong time.

It's a 3A or 5A linear regulator with voltage readout and alarm and electronic switchjack.

The 3A/5A difference comes from the selection of the actual regulator chip. The 3A units are a little bit cheaper but will shutdown sooner, the 5A ones are about double the price but give you more headroom. Ultimately at a -constant- current both will shut down from thermal effects before they run out of current ability. MOST R/C flight gear setups however will not be consuming 3 or 5A constantly - most are 0.3~1.0A average.

The back of this unit will have a 21x21mm heatsink attached, this will give us about 6'C/W cooling which means for the "typical" 50'C allowance that's almost 10W of dissipation - or 2V drop @ 5A (eg, 5V from 7V), so we've got a reasonable amount of margin there.

Hoping this one takes off in the market, as it's a nice compact unit, only 50 x 22mm ( 2 x 0.86" ) but should deliver solid power with very low noise.

Features:
  • Accepts A123, Lipo, Lilion
  • Extremely low loss switch MOSFET (0.075W loss)
  • Voltage readout via LED
  • Power good LED
  • Dual regulated power output headers, allows for cable failure or higher current delivery to RX or dual RXs
  • Audiable alarm available for Lipo version ( else generic alarm at ~6.2V input below which unit ceases to correctly regulate)
  • super compact, 50x22mm

Will take a single supply in from A123 or Lipo, provides dual outputs

Paul.

#2 PLD Nov 11, 2011 02:02 AM

PCB has been put in to production by the factory, probably 4~5 days and they'll ship it.

#3 Dan Zero-G Nov 11, 2011 11:44 AM

I am following this with great interest.

I would love to see some test results simulating F3J style loading with six powerhungry digital servo's and a 2s LiFE battery.

Peak load (launch, so servo's under heavy load or continuous stop start motion for all servo's - simulating butterfly deployment) 5A for 20 seconds (just for conservative check) and continuous load of around 300 mA and look for the voltage output and temperature changes (with no cooling, as we have to enclose them in our fuselages).

I love the dual output cables. I would hope the input cable is of a slightly larger cross section.

Also, please let us know the total weight ready to go.

:popcorn:

#4 PLD Nov 11, 2011 06:14 PM

Belginique,

It will be interesting to see how it holds up. There will be a 21x21mm pinhead heatsink on the back side (in part that is why there are so many holes under the regulator, so that it provides a better thermal conduction path when they're filled with solder).

The input leads will likely be 20AWG high-flex silicon wire, which should be ample for the capabilities of the system.

#5 PLD Nov 12, 2011 01:29 AM

Small update - I was looking around today at some cooling options and I've noticed I can get 20x20x10mm fans that only need 80mA of power and run off 5V, I was thinking that could be a nice addition for some people, they're not the cheapest fans around but it would add to the constant current handling ability (more than the increase that the 80mA for the fan would take).

#6 Mathelo Nov 12, 2011 07:16 AM

Quote:

Originally Posted by Belginique (Post 19847394)
I am following this with great interest.

I would love to see some test results simulating F3J style loading with six powerhungry digital servo's and a 2s LiFE battery.

Peak load (launch, so servo's under heavy load or continuous stop start motion for all servo's - simulating butterfly deployment) 5A for 20 seconds (just for conservative check) and continuous load of around 300 mA and look for the voltage output and temperature changes (with no cooling, as we have to enclose them in our fuselages).

I love the dual output cables. I would hope the input cable is of a slightly larger cross section.

Also, please let us know the total weight ready to go.

:popcorn:

What is the highest peak load that one could expect in this scenario?

Louis

#7 PLD Nov 12, 2011 07:28 AM

Actually looking forward to getting more data on things like this when I receive the circuit boards for the servo-logger.

#8 Dan Zero-G Nov 12, 2011 11:22 AM

Quote:

Originally Posted by Mathelo (Post 19854018)
What is the highest peak load that one could expect in this scenario?

Louis

That is a good question. There is a lot of theories around, but not much solid data. The best I could find was a post by Tuan, it is actual real-life data: http://www.rcgroups.com/forums/showp...&postcount=224

It shows flight loggin of an entire flight with a 2s LiFE pack unregulated. Peak loading during launch is about 3.5 Amps (but he takes 10 samples a second, so there could be some missing data). However, I think it means that a conservative estimate of max 5 Amps is not unrealistic.

Another thing is the stop / start current of digitals, but one would have to be very agressive on the sticks to have all servo's peaking at the same time (is my opinion anyway).

#9 PLD Nov 12, 2011 07:09 PM

That's some good data there. Considering that the 5A version will be able to handle 7.5~10A peaks it does give a reasonable safety overhead.

Interesting to note that for the rest of the flight the average is almost always under 1A.

#10 Mathelo Nov 13, 2011 07:11 AM

Quote:

Originally Posted by Belginique (Post 19855616)
That is a good question. There is a lot of theories around, but not much solid data. The best I could find was a post by Tuan, it is actual real-life data: http://www.rcgroups.com/forums/showp...&postcount=224

It shows flight loggin of an entire flight with a 2s LiFE pack unregulated. Peak loading during launch is about 3.5 Amps (but he takes 10 samples a second, so there could be some missing data). However, I think it means that a conservative estimate of max 5 Amps is not unrealistic.

Another thing is the stop / start current of digitals, but one would have to be very agressive on the sticks to have all servo's peaking at the same time (is my opinion anyway).

Thanks! That is helpful.

#11 PLD Nov 15, 2011 10:05 PM

Received notice that the PCBs have now been shipped... 2wks delivery now to wait :\

#12 PLD Nov 23, 2011 11:18 PM

The boards arrived today, they look good - will try get one built up tonight to test the performance.

Not really expecting any surprises here.

Paul.

#13 PLD Nov 24, 2011 05:49 AM

PLD-BAC-ESJ-3AR up and running
 
3 Attachment(s)
Some pictures of the newly built PLD-BAC-ESJ-3AR (yes, I know, the names are getting longer :( )

Size: 50 x 23 x 16mm
Weight: 9g
Battery input: 2S lipo or A123
Alarm point: 6.2V
Regulated output: 5.0V
Current: 3A

#14 Dan Zero-G Nov 24, 2011 11:40 AM

It would be great to see some test data giving it a hard time, simulating peak loading for 10-30 seconds of 5 amps then 10 minutes of 2 amps and then another minute of 5 amps for 1--30 seconds with no airflow to provide cooling and see how the voltage holds up and what the temperature fluctuations are.

Also simulated results for a spike of 10 amps if a couple of servo's would be seized up, just to see if it would survive such a beating.

It can't get much tougher than that for a F3J glider anyway.

It does look very pretty in any case!

#15 PLD Nov 24, 2011 07:01 PM

I think the task for this weekend is to build up a suitable load & test generator for these new regulators.


All times are GMT -5. The time now is 04:50 AM.