HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Dec 06, 2009, 11:47 PM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
BulletMaster,

Nothing overly yet though to be fair I've not really publicised it a lot (yet). From prior experience I like to get the first proto done and proven before making lots of noise.

I'm hoping to push this to the larger glider and glow market, especially with the dual-input feature. I also have to explore my heatsink options, while I have some already I don't think that they're going to suit the layout of the board in the best way.

At least with this PCB I had the sense to put the 4 mounting holes on it immediately

Paul.
PLD is offline Find More Posts by PLD
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 07, 2009, 02:01 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Okay, here's the preliminary test build. Already a couple of minor hiccups that I can pick up but nothing so severe that it'll stop me using the boards.

Right now I've only wired it up with 1 lipo input but the 2nd input is simply a JST lead away.

The built in audio alarm I'm thinking of putting to an external buzzer via a short flylead.

The way things are going, I think also I'll make a heatsink that covers most of the board area on the regulator side, that'll give me a nice 30x25mm finned area, quite a good deal of cooling from that.

Currently I'm turning it on/off via the little 2-pin jumper on the side of the LED/microcontroller, this will be replaced by a normal switch on a wired loom so it can be mounted elsewhere on the plane.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 07, 2009, 04:16 AM
Chillaxin dude!
BulletMaster's Avatar
Newcastle, NSW, Australia
Joined Apr 2003
4,813 Posts
looking good...
BulletMaster is offline Find More Posts by BulletMaster
Reply With Quote
Old Dec 07, 2009, 04:29 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Looks more and more like I'm going to have to go with some sort of thick flat aluminium or something for the heatsink, all the other heatsinks I can find (for BGA type chips) are just incredibly expensive.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 09, 2009, 04:44 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Changed things around a little bit today to get the BAC unit to be a bit more compact on the build.

Here's the pictures of the latest unit. The large capacitor folded over on the board is a 470uF low-ESR unit, more than ample to cope with the usual surge demands.

You'll notice also that it now shows both JST-2Slipo input leads and I've replaced the little jumper with a proper on/off switch.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 09, 2009, 04:48 AM
Chillaxin dude!
BulletMaster's Avatar
Newcastle, NSW, Australia
Joined Apr 2003
4,813 Posts
so no heat sink on this version? Or still to be fitted?
BulletMaster is offline Find More Posts by BulletMaster
Reply With Quote
Old Dec 09, 2009, 04:59 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Heatsink is yet to be found/fitted. In its current state it'll cope with about 1A of constant current but not a lot more.

What helps is that running the regulator at 5.3V and also passing the lipos via the protection diodes (schottky types) means that we end up dissipating less heat from the regulator compared to a normal 1A direct regulator (eg, the BAC-004's).

eg; @ 8.4V in, we'll drop 8.4 -0.4 -5.3 = 2.7V in the regulator, so for 1A that's 2.7W vs 3.4W for the 'traditional' setup.

@ 7.4V (the lower limit), we'll drop 7.4 -0.4 -5.3 = 1.7V, so 1.7W vs 2.4W for the 'traditional setup'

The chosen regulator has a 350~400mV dropout, so our voltage on the packs could drop as low as ~6.2V before regulation drops out, so there's plenty of headroom there.

Paul.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 09, 2009, 05:07 AM
HyperFlight Support
Neil Stainton's Avatar
United Kingdom, England, Stratford-upon-Avon
Joined Feb 2001
3,927 Posts
Hi Paul,

It is good to see you developing a new higher current regulator. Why not regulate to 5.5V? All R/C should withstand that voltage, and it will help with both power dissipation, and servo performance.

Neil.
Neil Stainton is offline Find More Posts by Neil Stainton
Reply With Quote
Old Dec 09, 2009, 05:09 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Neil,

I am thinking about pushing it to 5.5V, I agree with the reasons too. I just want to see however how well the 'real life' margin is at 3A when the battery is getting low, eg I don't want the regulator shutting down because it dropped below the drop-out limit.

Paul.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 09, 2009, 05:17 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Good news, I found a heatsink that's very nearly perfect, it's about 1mm too long but it's a good size other than that, the hard/downside bit is that I have to bond it to the regulator using 2-part epoxy solution which means that it'll be a pain to service if anything goes wrong (though the epoxy will breakdown beyond 150'C).

The sink offers 8.5~10'C/W dissipation, so that means at 3A at full voltage and say at 5.5V regulated, that'll require us to put 7.5W into the heatisnk , so, 65~75'C above ambient, while that is hot it is also still below 150'C, even if you consider an ambient of 50'C in the airframe.

The glue to bond the heatsinks on is also stupidly expensive ($30 for 5mL). Ah well, such is the curse of these small electronics applications.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 09, 2009, 08:07 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
I was thinking, another thing I could use is simply aluminium bar (which I get here in 2.4m lengths), though again, bonding it to the regulator is in itself half of the problem. If I used bare copper it'd be easy, one could just solder it
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 09, 2009, 09:22 PM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Heatsinks and bonding solutions have been ordered, with any luck we'll see them arrive tomorrow around this time.

I managed to locate some reasonably priced double-sided thermal bonding tape, it's only about 0.5mm thick and has the same transfer characteristics (~0.8'C/W) as the epoxy solutions but it is obviously a lot easier to handle (one hopes).

Paul.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 10, 2009, 09:15 PM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Okay, here's the unit with the heatsink installed. It would be nice to have had it a bit wider (to match the other side of the board) but this is still quite a decent setup.

I'll probably need to put a blob of RTV onto the other end of the heatsink just to provide some level of mechanical support (or perhaps a slice of double-sided adhesive foam).
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 28, 2009, 04:26 AM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
Finally got some thermal testing done of the BAC unit and it's not quite as good as I want at the moment.

With an ambient of 29'C, no airflow, drawing 2.2A @ 5.35V into a 2.5R dummy load, the heatsink gets up to 85'C and the regulator goes into "thermal shutdown" mode after a 2 minutes. At this level the heatsink is trying to dump off about 5.8W of heat.

If positioned in the airflow the setup would tolerate a higher constant current.

If I raise the voltage up to 5.5V then the heat dissipated will be a bit less but not significantly so ( P = IV, so P1 = 2.2 x 2.65 = 5.8W, vs P2 = 2.2 x 2.5V = 5.5W )

Not that there's anything profoundly unusual here to be honest, rarely can you do the rated current on regulators constantly without quite large heatsinks.

The realistic thing to do here is write up the specifications much like I have with the smaller BACs,

Constant rating (eg 1.5A), 1 minute rating (eg, 2A) and 10~30 second rating (eg 3A).

Paul.
PLD is offline Find More Posts by PLD
Reply With Quote
Old Dec 28, 2009, 10:19 PM
PLD
Magicsmoke maker
PLD's Avatar
Australia, QLD, Charters Towers
Joined Dec 2003
17,404 Posts
New tests with 3.3R load

Made up a 3.33R load today and the regulator was fine to run that non-stop.

Current was about 1.6A, temperature got to ~70'C in 30'C ambient with high humidity (light drizzle outside).

I'd consider this about the 'limit' of the permanent load rating.

If you get airflow over the regulator, even a light breeze will drop the temperature by about 10'C.
PLD is offline Find More Posts by PLD
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
A question for those who use high cell counts as opposed to higher current Sal C Power Systems 14 Feb 25, 2006 06:11 PM
Charge identical packs in parallel at higher current? lynnpreston Batteries and Chargers 3 Apr 14, 2004 08:56 AM
Question Proportionaly switching to a higher current? infrared Scratchbuilt Indoor and Micro Models 7 Mar 30, 2004 09:56 PM
Higher current output LiPoly chargers? RD Blakeslee Batteries and Chargers 3 Oct 28, 2002 04:20 PM
lightweight but higher current cells? t-turley Batteries and Chargers 6 Jul 26, 2002 05:40 AM