I tested a lot of Castle products over the years. Their BECs are actually among the best you can get. A BEC like this is meant for high powered airplanes or helicopters which would have ridiculously high airflow and it's generally assumed that you will mount it as such. It's also worth mentioning that the dominant load on these BECs is servos so it's virtually guaranteed that peak loading will never happen for more than a few seconds at a time. Check out their ripple and transient stability if it's anything like their entry level BEC it should be good.
Just because there are more expensive products doesn't mean that one isn't expensive for what it does. It doesn't look bad, but it's still just an off-the-shelf DC-DC converter with a "meh" build quality. Oh, and regarding Western Robotics, paying more than 100USD for a general purpose low voltage DC-DC converter is getting into the bullshit region, at least for me, they charge 50USD for a 5A BEC, ffs...
scott firman To be honest it's not that hard, at low voltages/currents you can just get away with a linear regulator with an heatsink and a couple of capacitors. Even with switching regulators, most manufacturers have an application note telling you how it should be done for each driver IC, like this: www.st.com/content/ccc/resource/technical/document/application_note/75/1c/c6/24/bc/a5/4b/94/CD00043744.pdf/files/CD00043744.pdf/jcr:content/translations/en.CD00043744.pdf In a pinch you can even use a good quality car phone charger, they come rated at 2-3 Amp or more and most can handle up to 35V.
25:25 - It likely failed at the 15A output because Dave's supply was only outputting 15.2V AT THE POWER SUPPLY. While relatively small there will be a voltage drop over those leads at the required current (7A+). Plus the manufacture listed 16V not 15V as the required input voltage. If it only needed 15.2V they would have marked it 15V and not 16V.
In real world usage, this BEC is used (and aimed) to power only the servos that control the various control surfaces of a model airplane (ailerons, elevator, rudder etc). It's not used to drive the main motor of the airplane. That means that it will probably never be maxed out in power draw if only maybe for a couple of seconds. And generally people that fly RC planes like myself we always use a BEC that can provide more power than all of the servos would require combined. Having said that this BEC wouldn't even get warm to the touch when in flight, even if you do crazy aerobatics with the plane which wouldnt last more than 10-15minutes of flight anyways before you need to land. So this is a good BEC and serves its purpose just fine.
The reason it goes down to those connectors with the thinner wire is that those are the standard connector for a BEC, servo, or ESC in the RC industry and they are limited to the power that can go through the pins. More advanced users (many of which buy Castle products) can just chop off the connector and wire into whatever needs power through the BEC.
+EEVblog usually, no, and definitely not continuously. the main power drain for these are the servos that control the control surfaces. As you can imagine, the power usage will be highly fluctuating. The max power drain probably only will last for a couple of seconds. Also, a normal flight time for an electric plane is less than 10 minutes, less if you fly aggressively which is when you really strain this.
What Joni above this said, for the most part it wouldn't' be running anywhere near maximum current rating, and if it does it would only be momentary. The only thing that I can think of that would cause a consistent current drain would be lights (LED's) and very small fans (for cooling). So for the sake of it, That unit seems to be rather decent considering!
I also agree with Joni. Even in a highly acrobatic airplane the servo loading is almost never going to be used near 100% it's maximum rated current. The main fear is never to have the voltage drop below minimum for the R/C receiver. (Known as a brown out causing loss of power to the receiver and thus loss of control.) You can live if your servos slow down from some loss in voltage, but once the receiver is below it's threshold even for a instant it will cause a reboot and then a re-sync to lock on to the correct Chanel. This time delay can be several seconds and deadly to an aircraft.
Hey Dave. I'm one of Keith's aeromodeller friends and just wanted to let you know, YOU ROCK! Thank you so much for making this video. This was one of the most informative hobby-related things I've watched in quite a while :)
When taking those thermal measurements, they should have been done with the PCB in the actual plastic case, as from my perspective, it obstructs the heatsink somewhat.
Good point. It blocks the heat sink a bit but even more it totally blocks any air flow directly over the FETs on the component side. Dave should have used a thermocouple glued to a FET and put the covers back on (little hole for the wire of course) to get a more accurate result. But it was only a rough test really.
sweet video, I've always wanted to know how close to spec they are! As others have said, castle are a big name brand. a long time ago they publicly stated that they were hesitant to put any ratings on anything because they would seem completely uncompetitive when compared with the absolutely insane and exaggerated ratings used by everyone in the rc industry. they started to add some ratings, that are still a fair bit lower than directly competitive products, but pretty much everyone knows they're more accurate. I think they even have a standard for what continuous means in the rc world. also in the manual it's stated that the little plugs are only rated to 5A each and to add more or different connectors if you need more current.
This is over rated but giving Model aircraft dont draw constant current is actually really good , also planes and helicopters give some airflow since they are flying
+Rcmodellbau_DE Is it? The person who sent it in had it programmed for lower current for his set up and how many cell battery packs he runs. Dave pushed it last for how it was programmed and that's why it fell off. If it was reprogrammed then it would have done what it claims to. But the programing of these ESC or BEC is were I see most people fail and burn up the electronics in RC.
+Gadget Review (GRV) Most people dont even know how to properly program or set these things up in the first place. Remember too,the rating are listed at Peak ratings,only designed for short bursts,not prolonged like Dave did,that was totally uncharacteristic of how anyone would fly,or load up the BEC. Besides,I found out the hard way,its the transmitter that tends to fail on extended stick usage,power consumption.
The price is pretty normal.. remember 90% of becs are rated for 2amps max and use linear approach (7805). That said.. Castle creations is Apple of rcs. Overengineered and overpriced in most cases.
I'm still watching so perhaps you address this later, but I'm pretty sure you're going to miss an important point that would be obvious to an RC person... wind! (Not your fault! You're an EE blog, not an RC blog...) When in use for its intended purpose (ie: model aircraft) the BEC can be subject to a quite fierce airflow, depending on placement. RC people regularly run equipment at power levels that would fail on the ground. Similar to the way professional cyclists sometimes overheat when they're on stationary bikes. An RC person who expected to run this "hot" would put it right in the propwash or even force some airflow through the fuselage. To replicate that, you'd need something like a leafblower. I'm impressed it went into thermal shutdown rather than just releasing it's magic smoke. In RC terms, that's basically "bulletproof". (Now if they could just make them tree-proof!) [edit] Five miles an hour? *giggle* Any plane that needs a BEC that big and is only going 5mph is probably a miniature model of the Spruce Goose. Your intuition towards the end that "with some good airflow this might work" was right on track. Model planes can go 60-100kph, which is the trick.
I did model aircraft for a few years. Most of the models either have space behind the cowl for your speed control or regulator, or they have a largish hole there that feeds air into the plane with an exit hole near the tail. The mounting point for the electronics would always be in the airflow path. Also, you'd get far more than 5mph airflow there. After a long flight, my temp sensor would usually show the heat sink sitting around 60 to 62C.
It's interesting about the output connectors. Those are standard servo connectors and have been in use in rc for many many years. When they started being used the power handling requirements were a lot lower than they are today.
I'm watching the video..., Dave talks about the temperature from the power dissipation of the heat sink and I'm like "huh. wouldn't it be great if he used a thermal imaging camera to evaluate this...". Then he pulls out a Flir E8 - something which i could only dream of with my PhD budget (we're currently "borrowing" a Flir i7 from another department - less spec) to show it on video.I wish I could subscribe twice! thanks for the quality vids Dave! great stuff! :P
Castle is generally reguarded as the better quality BEC untis available out there. I see way too many guys bury them so they get inefficent air flow. The result is generally a brown out,this happends durning a manuver that requires more control surface imput,that inturn puts more of a load on the BEC,resulting in a brown out,loss of control and crash. Then its always blamed on servo's of course.I always use a ESC with a switching BEC seperate from the ESC so if the amp rating is drawing too much power,I dont lose power to my control surfaces.I havent experienced a problem since doing that. Even if my ESC goes up in smoke,I can still land my plane.
+KpilotRCHelis No one is going to dump 100 bucks into a BEC on a foam plane. When you start talking $1000 dollars plus,then it only makes sense. Thats like Transmitters,I recently experienced the disappointment of a Spektrum DX6I not being able to handle prolonged imput which resulted in a brown out. This happened twice on two craft. I had read these things couldnt handle excessive stick movement and found it to be true,twice on two different planes. I dont have unlimited funds for my equipment and dont see any reason it shouldnt have been able to handle imput that was created because the unit was not responding properly in the first place. I have noticed that,yes you get what you pay for in RC but in the same sense,you shouldnt have to dump even more money into the equipment because the electronics are crap to begin with. Its like buying a new car for 14 k and having it constantly breaking down because of inferior parts as opposed to paying $150k for a new car and having better parts,but still having to invest in better parts. eather way,you shouldnt have to keep dumping money into it because they used crappy parts.
Awesome testing. As a big fan of Castle Creations products it's nice to see something they make put to the test. Pretty crazy how much a little airflow affects the temperature.
Sweet, I think you confirmed that this BEC is pretty useful! Failing to hit the 15v current spec is disappointing but it looks like it can output 8A+ at all input voltages that it's rated for, and if you pull that continuously you just have to make sure you have airflow over it. Considering that most RC vehicles this would be in travel far faster than 5mph it shouldn't be hard to get that much airflow over it, and 8A is quite a lot for most applications (which probably peak at a lot less than 8A). Should be no surprise to RC people as Castle Creations are one of the big RC electronics manufacturers. Nice to have confirmation though!! Thanks!
Awesome to see these on your'e channel. I bought a similar thing rated 8A, but the datasheet on the chip was only capable of 3. Ive also had one on my miniquad, that failed sending 14V through all the 5V electrics, ruining $100 worth of flight controller, receiver, camera and vtx.
The logo on those two cap cans is the same one on the solid polymer caps I bought for a Radeon HD 5870 GPU card six years back. For the life of me, I can't remember the brand, but I remember it was a well-known name. For what it's worth, I only just retired that same GPU card last week and it still works.
LOL. Ya SOME people just have to comment before watching the whole video... and two commenters already had to comment on their comments because it was covered later in the video! Like just watch the WHOLE video first and save everyone the hassle of reading your incorrect comment! Some never learn!
Sometimes faulty BECs cause RC radio signal loss. Can you please have a look at the RF spectrum... :) Maybe use a LISN and show us the conducted emissions. :) Or some nearfeild probes and show us the noise band.
Can i just point out that the transistors is NOT what is going to cause thermal problems in a regulator like this!? Dave implies that the transistors will be the issue, but there is nearly no voltage across them when they are conducting, so also very low power dissipation. The diode is what will cause your thermal headache. Depending on duty cycle of cause, but often the diode dissipates much more power. That is one reason why synchronous rectification in DC-DC converters is so awesome.
Castle speed controls (with built-in BEC) flawlessly power most of my RC trucks and helicopters. The same small piece of software from Castle (Castle-Link) lets me adjust parameters on all of them, allowing programming for almost any application or condition. As always, thanks for your unbiased reviews. Keep up the great work, and remember... Don't turn it on - Take it apart !!
Been re-watching some of Dave's old vids, geez they just are soooo good. Watched the 'Murphy U-Current 'one, and, well watch it! and if you don't remember from the first time, try and do your own detective work. Seriously brilliant.
You may have mentioned this but typically no one ever pushes their electronics to the peak as what is listed on that BEC. Those numbers are all PEAK numbers,not actual rates your Equipment would be putting out. I am talking about running full throttle,cycling landing gear retracts,full navigation lighting,operating servos at full capacity,flaps fully extended or being retracted. Usually a failure occures just after take off or landing because of those reasons. Expect a Transmitter brownout before that BEC to fail though.
with high permeability ferrites curie temperatures can be as low 120°C so that can be a problem as well, ceramic capacitors also dont really like high temperatures as do electrolytics of course
Hi Dave, 48.9W is Power consumed by the BK Electronic Load, Regulator is producing 52.5W , Delta around 3W is the Joule effect produced in the cable , gone in heating the cables. on my point of view its not an error . you doing a great job ! Ali.
Nice one Dave, thanks! A friend of mine uses a BEC that looks very much the same heatsink in his aeroplane. As his systems engineer :-) I had him try a couple different geometries to mount the BEC in a "prototyping fixture" I jury rigged. Best performance so far has heatsink fins getting air "scooped" & directed over fins
The heatsink data sheet calls for a 1" square heatsink, 25x25mm. Now I'm not confident of the scaling, but the heatsink on the device looks significantly larger than that to me - between 30 and 40 mm square.
Dave, Castle Creations makes some of the better reliable BEC units in the RC Industry. I wouldn't worry about the lifespan of the caps. Most likely the failure of the BEC would be due to a crash. :-)
crashes are usually due to failures such as this. Most of us that have flown for years,rarely crash and usually its on landing due to cross wind landings. I havent had a crash in quite a while, the last one was caused by servo failure. I dont expect electronics to hold up for longer periods of time and generally most electric motor planes only fly 15 minutes tops while gas is 25-30 minutes at best. Like I said, most guys I fly with are pro's and generally dont crash. when you invest 3 to 5 k or more into a plane or heli,you usually dont go with cheap BEC's or U BEC's. I use An ESC and seperate BEC. I see too many planes not having enough air space to properly cool these types of BEC's.Happy flying!
I think it's safe to say BEC's are rarely run at their absolute max rating continously anyway. It's a varying load - I can't think of many situations where a heli or plane will be asking for maximum juice during an entire flight. Unless you're Alan Szabo.
even so i flight helis and planes as well. but I never had the same curious though about all my BEC and speed controls like Dave is explained to us and I know that castle creation is one of the top brands out.
Was liking the look of your Flir thermal imaging cam Dave, until I found the price!!! £4000 in the UK!! I suppose it was a freeby then! On using the voltage reg, presumably it would be mounted exposed to airflow created by prop wash? In which case, plenty of air flow!
Those output caps are Chemi-Con PS_ series (not sure of the last letter, series name couldn't fit on the top there) solid polymer electrolytics - i wouldn't be concerned with those kicking the bucket anytime soon. Besides, their tops are even "outside" so they get some cooling from the airflow.
I think the "continious" means 1 batteryload. most batteries only stand for about 10 minutes. when you change the battery its always reconommed to wait until the motor and speed controller are cooled down.
All your calculations I believe were based upon 20* ambient temperature. In a hotter climate like 40* or maybe 45* ambient I just wonder if the converter will hold its regulation. Just a thought
I think it's safe to say if Dave built any kind of RC thing it would be overbuilt in so many ways and nearly impossible to kill. :P Overkill is my favorite kind of kill.
I think the ratings are nothing to do with the BEC output. A BEC is designed to run from the flight battery in an electric model that drives the motor, and provide a low voltage (typically 4.8v) to power the receiver and servos, which typically don't require a huge amount of current. Before BEC's came along, most people had to use a separate battery pack (often 4 x AA ni-cads) to power the receiver and servos. I doubt that a BEC would be expected to supply 8 amps or more !
I fly electric heli's up to 90 size (250+ amps into the motors). I use this type of controller all the time and they are normally used to power radio and servo's etc that create short term peak loads rather than continuous maximum loads. If you want to run them at full power continuously then yes you need to put them in air flow and they work fine. In other words they work fine for the intended purpose.
Do take into account that this device is intended for use on an airplane/heli with plenty of airflow. If you're at max load and only going 5mph, something is really wrong. I think I would opt to operate it with the case off.
They are usually mounted inside the airframe, though, so 2~3 m/s is reasonable enough and it had the the added bonus of being tested without the airflow restriction of the plastic case inside a lab with air conditioning.
Ben The Shrubber He knows, he was arguing that fan wasn't good enough (hence the 5mph number), I was arguing it is, as in most cases the BEC is mounted inside the airframe with limited venting holes.
ohh.. its just a bec... 8amp would be massive for servos... prob way overspecced product... or for big planes. Normally the BEC is with the speed controller, that would be more interesting viewing, seeing if it can send 120 amp continous.. some of the old boys like nicd batteries for BECs. The new coreless servos often with their wide voltage op ranges can run straight off the lipo.. with good torque characteristics
www.hobbyking.com/hobbyking/store/__20555__Turnigy_dlux_100A_SBEC_Brushless_Speed_Controller_w_Data_Logging_AUS_Warehouse_.html?strSearch=turnigy%20bec get that up ya
Great testing there Dave. Never thought to stress test my BECs this way as I have the same PSU and a Maynuo EL (which looks to be a ripoff of your BK unit). As many have said the Castle creations units are certainly one of the better brands out there for this sort of gear and they're not really expensive as you mentioned. I have a couple of those CC BECs but have never really stressed them - they are way over sized for even my large RC heli's. For reliability I've now removed the BECs and gone back to a dedicated control battery (2S Lipo). Some high performance servos and gyros ( for Heli's anyway) now directly run at the higher 7.4volts (as opposed to the traditional 5 volts) of a 2S battery so the BEC can be totally eliminated in my case. One less thing to go wrong IMO. Once again - great video.
Can't imagine the caps being the limiting factor for the device's lifetime. Would have thought its end would be more likely getting augured into the ground...
Cool vid. One of the more interesting ones I've watched in a while :-) Love the display on that B+K load. I have one of their bench multimeters (5491B) with a similar display, it's a joy to work with in a wide range of ambient lighting levels
What about if it was running at 16 volts exactly instead of 15.2, would it manage to output 15 Amps or would the 0.8 Volts not be enough to cause a problem?
If you could raise the output voltage from 6.5V to something more like 10V, it might very well be able to supply 15A. I wouldn't mind seeing more on this item if you find time :)
That thing could have used a synchronous rectifier FET, would have cut losses in half if not more, especially at high voltages deltas where the free-wheeling diode/FET carries the bulk of load current during the 80+% off-time.
The usual overstating of specs for most hobby products. What would help this survive is probably the fact that most of the load would come from the servos moving the control surfaces of the plane so only in short bursts as it maneuvers. Not much load when all the servos are neutral(eg flying straight).
I still don't understand why you would need that much power for a BEC. On my model car it was just a way to power the radio or other devices that would have minimal draw, eliminating the four "AA" batteries in series, unlike the motor which runs directly off the "Large main" battery or speed control. Other then the motor what other high drain devices would use this on a plane car or helicopter? At first I thought it was a speed control looking at the beefy wires.
Four high voltage standard servos producing 25kg-cm of torque and 0.05s/60deg will easily draw 2A during travel with no load on the output and much more peaks. Helis use four, and huge scale airplanes can use up to 8. That's some serious power. Cars are nothing compared to aircraft.
With model airplanes every gram counts so massive block of aluminium only shortens the battery life (which is 10-15 minutes anyway) or renders it unable to fly if it is too heavy.
Anodized aluminum probably has a different emissivity than the default setting in your flir camera. Did you take that into account for your temperature measurement ?
8:28 Incorrect, polymer capacitors have lifetime ratings all over the place, just like electrolytics. But yes, there are some large vent-less electrolytics out there. Specifically, the Sacon FZ series, which are known to go off like gunshots!
I tested a lot of Castle products over the years. Their BECs are actually among the best you can get. A BEC like this is meant for high powered airplanes or helicopters which would have ridiculously high airflow and it's generally assumed that you will mount it as such. It's also worth mentioning that the dominant load on these BECs is servos so it's virtually guaranteed that peak loading will never happen for more than a few seconds at a time. Check out their ripple and transient stability if it's anything like their entry level BEC it should be good.
it seems pretty well made and though out, seems like good parts too
Yeah if you want best..talk to NASA, they might surprise you
@@jhonnyginting8985 Yeah, NASA uses a LOT of Castle products
"This thing's cheap, it's like $38 or something which is apparently cheap for one of these things"
Nope that's expensive af.
Remember it's not a 4s 5A BEC, it's an high voltage high current one from a reasonably well established brand name.
Lol you think that is expensive?
Look up Western Robotics BEC. Those are the top in the RC world and are much more beefy.
Just because there are more expensive products doesn't mean that one isn't expensive for what it does. It doesn't look bad, but it's still just an off-the-shelf DC-DC converter with a "meh" build quality. Oh, and regarding Western Robotics, paying more than 100USD for a general purpose low voltage DC-DC converter is getting into the bullshit region, at least for me, they charge 50USD for a 5A BEC, ffs...
+TrickyNekro You build your own BEC's? WOW! good for you.
scott firman To be honest it's not that hard, at low voltages/currents you can just get away with a linear regulator with an heatsink and a couple of capacitors. Even with switching regulators, most manufacturers have an application note telling you how it should be done for each driver IC, like this: www.st.com/content/ccc/resource/technical/document/application_note/75/1c/c6/24/bc/a5/4b/94/CD00043744.pdf/files/CD00043744.pdf/jcr:content/translations/en.CD00043744.pdf
In a pinch you can even use a good quality car phone charger, they come rated at 2-3 Amp or more and most can handle up to 35V.
25:25 - It likely failed at the 15A output because Dave's supply was only outputting 15.2V AT THE POWER SUPPLY. While relatively small there will be a voltage drop over those leads at the required current (7A+). Plus the manufacture listed 16V not 15V as the required input voltage. If it only needed 15.2V they would have marked it 15V and not 16V.
In real world usage, this BEC is used (and aimed) to power only the servos that control the various control surfaces of a model airplane (ailerons, elevator, rudder etc). It's not used to drive the main motor of the airplane. That means that it will probably never be maxed out in power draw if only maybe for a couple of seconds. And generally people that fly RC planes like myself we always use a BEC that can provide more power than all of the servos would require combined. Having said that this BEC wouldn't even get warm to the touch when in flight, even if you do crazy aerobatics with the plane which wouldnt last more than 10-15minutes of flight anyways before you need to land. So this is a good BEC and serves its purpose just fine.
The reason it goes down to those connectors with the thinner wire is that those are the standard connector for a BEC, servo, or ESC in the RC industry and they are limited to the power that can go through the pins. More advanced users (many of which buy Castle products) can just chop off the connector and wire into whatever needs power through the BEC.
Glad you actually tested this thing Dave. I fly rc planes, and there's so much false advertising when it comes to the electronics!
To fully test it and characterise it though would take days. I have no idea if it will ever be used near it's maximum rated current?
+EEVblog usually, no, and definitely not continuously. the main power drain for these are the servos that control the control surfaces. As you can imagine, the power usage will be highly fluctuating. The max power drain probably only will last for a couple of seconds.
Also, a normal flight time for an electric plane is less than 10 minutes, less if you fly aggressively which is when you really strain this.
What Joni above this said, for the most part it wouldn't' be running anywhere near maximum current rating, and if it does it would only be momentary. The only thing that I can think of that would cause a consistent current drain would be lights (LED's) and very small fans (for cooling). So for the sake of it, That unit seems to be rather decent considering!
And also there is usually some airflow due to the craft moving along.
I also agree with Joni. Even in a highly acrobatic airplane the servo loading is almost never going to be used near 100% it's maximum rated current. The main fear is never to have the voltage drop below minimum for the R/C receiver. (Known as a brown out causing loss of power to the receiver and thus loss of control.) You can live if your servos slow down from some loss in voltage, but once the receiver is below it's threshold even for a instant it will cause a reboot and then a re-sync to lock on to the correct Chanel. This time delay can be several seconds and deadly to an aircraft.
Hey Dave. I'm one of Keith's aeromodeller friends and just wanted to let you know, YOU ROCK! Thank you so much for making this video. This was one of the most informative hobby-related things I've watched in quite a while :)
shouldn't you have put a fan on it? I would imagine in flight it would dissipate heat pretty well and it only has to last 30 minutes max.
He did put a fan on it.
Yeah I saw that. I commented before I finished the video.
I guess you didn't watch the video!
+Bottleworksnet I did
Did you not watch the video?
When taking those thermal measurements, they should have been done with the PCB in the actual plastic case, as from my perspective, it obstructs the heatsink somewhat.
Good point. It blocks the heat sink a bit but even more it totally blocks any air flow directly over the FETs on the component side. Dave should have used a thermocouple glued to a FET and put the covers back on (little hole for the wire of course) to get a more accurate result. But it was only a rough test really.
sweet video, I've always wanted to know how close to spec they are! As others have said, castle are a big name brand. a long time ago they publicly stated that they were hesitant to put any ratings on anything because they would seem completely uncompetitive when compared with the absolutely insane and exaggerated ratings used by everyone in the rc industry. they started to add some ratings, that are still a fair bit lower than directly competitive products, but pretty much everyone knows they're more accurate. I think they even have a standard for what continuous means in the rc world. also in the manual it's stated that the little plugs are only rated to 5A each and to add more or different connectors if you need more current.
Mate your videos are ALWAYS longer than you thought they would be. :)
This is over rated but giving Model aircraft dont draw constant current is actually really good , also planes and helicopters give some airflow since they are flying
Also it can output up to 12.5v max
But it says that the BEC is capable of a CONSTANT current draw of the said amps. Clearly false advertisement.
The BEC is rated with 3mph airflow over the heatsink ...
+Rcmodellbau_DE Is it? The person who sent it in had it programmed for lower current for his set up and how many cell battery packs he runs. Dave pushed it last for how it was programmed and that's why it fell off. If it was reprogrammed then it would have done what it claims to. But the programing of these ESC or BEC is were I see most people fail and burn up the electronics in RC.
+Gadget Review (GRV) Most people dont even know how to properly program or set these things up in the first place. Remember too,the rating are listed at Peak ratings,only designed for short bursts,not prolonged like Dave did,that was totally uncharacteristic of how anyone would fly,or load up the BEC. Besides,I found out the hard way,its the transmitter that tends to fail on extended stick usage,power consumption.
cheap? castle is brand name in the rc world. A cheap bec is less than 10$ :)
Ok.
I use few of these since 2008 and none broke but they do browout / power cycle under too much load. I wish I could make my own.
This BEC is "inexpensive" cheap can sometimes imply poor quality
The price is pretty normal.. remember 90% of becs are rated for 2amps max and use linear approach (7805). That said.. Castle creations is Apple of rcs. Overengineered and overpriced in most cases.
Western Robotics 14s BEC is $100 and much more beefy
I'm still watching so perhaps you address this later, but I'm pretty sure you're going to miss an important point that would be obvious to an RC person... wind! (Not your fault! You're an EE blog, not an RC blog...)
When in use for its intended purpose (ie: model aircraft) the BEC can be subject to a quite fierce airflow, depending on placement. RC people regularly run equipment at power levels that would fail on the ground. Similar to the way professional cyclists sometimes overheat when they're on stationary bikes.
An RC person who expected to run this "hot" would put it right in the propwash or even force some airflow through the fuselage. To replicate that, you'd need something like a leafblower.
I'm impressed it went into thermal shutdown rather than just releasing it's magic smoke. In RC terms, that's basically "bulletproof". (Now if they could just make them tree-proof!)
[edit] Five miles an hour? *giggle* Any plane that needs a BEC that big and is only going 5mph is probably a miniature model of the Spruce Goose. Your intuition towards the end that "with some good airflow this might work" was right on track. Model planes can go 60-100kph, which is the trick.
yea especially since the prop wash is way over 5 mph wind
I did model aircraft for a few years. Most of the models either have space behind the cowl for your speed control or regulator, or they have a largish hole there that feeds air into the plane with an exit hole near the tail. The mounting point for the electronics would always be in the airflow path. Also, you'd get far more than 5mph airflow there. After a long flight, my temp sensor would usually show the heat sink sitting around 60 to 62C.
It's interesting about the output connectors. Those are standard servo connectors and have been in use in rc for many many years. When they started being used the power handling requirements were a lot lower than they are today.
Always a good time watching your electrical tear downs. Cheers!
I'm watching the video..., Dave talks about the temperature from the power dissipation of the heat sink and I'm like "huh. wouldn't it be great if he used a thermal imaging camera to evaluate this...". Then he pulls out a Flir E8 - something which i could only dream of with my PhD budget (we're currently "borrowing" a Flir i7 from another department - less spec) to show it on video.I wish I could subscribe twice! thanks for the quality vids Dave! great stuff!
:P
Castle is generally reguarded as the better quality BEC untis available out there. I see way too many guys bury them so they get inefficent air flow. The result is generally a brown out,this happends durning a manuver that requires more control surface imput,that inturn puts more of a load on the BEC,resulting in a brown out,loss of control and crash. Then its always blamed on servo's of course.I always use a ESC with a switching BEC seperate from the ESC so if the amp rating is drawing too much power,I dont lose power to my control surfaces.I havent experienced a problem since doing that. Even if my ESC goes up in smoke,I can still land my plane.
The Castle BEC is not really known to be reliable. If you want something quality, Western Robotics are the best and most robust you can get.
The Castle BEC is not really known to be reliable. If you want something quality, Western Robotics are the best and most robust you can get.
JETI all the way :P i fly large 1/3rd scale aircraft i wouldnt trust my $5000 plane to anything else :P
+warmfreeze You say that, but do you know why. Why is your Jeti gear the only thing that can be trusted? Because it's expensive?
+KpilotRCHelis No one is going to dump 100 bucks into a BEC on a foam plane. When you start talking $1000 dollars plus,then it only makes sense. Thats like Transmitters,I recently experienced the disappointment of a Spektrum DX6I not being able to handle prolonged imput which resulted in a brown out. This happened twice on two craft. I had read these things couldnt handle excessive stick movement and found it to be true,twice on two different planes. I dont have unlimited funds for my equipment and dont see any reason it shouldnt have been able to handle imput that was created because the unit was not responding properly in the first place. I have noticed that,yes you get what you pay for in RC but in the same sense,you shouldnt have to dump even more money into the equipment because the electronics are crap to begin with. Its like buying a new car for 14 k and having it constantly breaking down because of inferior parts as opposed to paying $150k for a new car and having better parts,but still having to invest in better parts. eather way,you shouldnt have to keep dumping money into it because they used crappy parts.
You should test more RC components like ESC's and things. That's what got me interested in electronics
That's why these have earned the nick-name 'Castle Cremations'.
Awesome testing. As a big fan of Castle Creations products it's nice to see something they make put to the test. Pretty crazy how much a little airflow affects the temperature.
For anyone who does anything RC. Those twin output cables are designed for signal and power to servos.
Sweet, I think you confirmed that this BEC is pretty useful! Failing to hit the 15v current spec is disappointing but it looks like it can output 8A+ at all input voltages that it's rated for, and if you pull that continuously you just have to make sure you have airflow over it. Considering that most RC vehicles this would be in travel far faster than 5mph it shouldn't be hard to get that much airflow over it, and 8A is quite a lot for most applications (which probably peak at a lot less than 8A).
Should be no surprise to RC people as Castle Creations are one of the big RC electronics manufacturers. Nice to have confirmation though!!
Thanks!
Awesome to see these on your'e channel. I bought a similar thing rated 8A, but the datasheet on the chip was only capable of 3.
Ive also had one on my miniquad, that failed sending 14V through all the 5V electrics, ruining $100 worth of flight controller, receiver, camera and vtx.
Dave, for further testing at all ranges you would need a Castle link USB adapter.
The logo on those two cap cans is the same one on the solid polymer caps I bought for a Radeon HD 5870 GPU card six years back. For the life of me, I can't remember the brand, but I remember it was a well-known name. For what it's worth, I only just retired that same GPU card last week and it still works.
cool video Dave. loved it and the nice part that the test didn't bread the DUT. very cool.
Video is 31 mins long and yet just minutes after upload and there is comments and upvotes and downvotes, geez people.
LOL. Ya SOME people just have to comment before watching the whole video... and two commenters already had to comment on their comments because it was covered later in the video! Like just watch the WHOLE video first and save everyone the hassle of reading your incorrect comment! Some never learn!
Patron supporters get to watch and obviously comment a few hours before the video is switched to publicly visible
Sometimes faulty BECs cause RC radio signal loss. Can you please have a look at the RF spectrum... :)
Maybe use a LISN and show us the conducted emissions. :)
Or some nearfeild probes and show us the noise band.
Can i just point out that the transistors is NOT what is going to cause thermal problems in a regulator like this!? Dave implies that the transistors will be the issue, but there is nearly no voltage across them when they are conducting, so also very low power dissipation. The diode is what will cause your thermal headache. Depending on duty cycle of cause, but often the diode dissipates much more power. That is one reason why synchronous rectification in DC-DC converters is so awesome.
didnt they have two diodes probably in parallel?
Castle speed controls (with built-in BEC) flawlessly power most of my RC trucks and helicopters. The same small piece of software from Castle (Castle-Link) lets me adjust parameters on all of them, allowing programming for almost any application or condition. As always, thanks for your unbiased reviews. Keep up the great work, and remember... Don't turn it on - Take it apart !!
As always, a thorough and thoroughly enjoyable video. Thanks a million!
Been re-watching some of Dave's old vids, geez they just are soooo good. Watched the 'Murphy U-Current 'one, and, well watch it! and if you don't remember from the first time, try and do your own detective work. Seriously brilliant.
You may have mentioned this but typically no one ever pushes their electronics to the peak as what is listed on that BEC. Those numbers are all PEAK numbers,not actual rates your Equipment would be putting out. I am talking about running full throttle,cycling landing gear retracts,full navigation lighting,operating servos at full capacity,flaps fully extended or being retracted. Usually a failure occures just after take off or landing because of those reasons. Expect a Transmitter brownout before that BEC to fail though.
with high permeability ferrites curie temperatures can be as low 120°C so that can be a problem as well, ceramic capacitors also dont really like high temperatures as do electrolytics of course
as an RC guy I really like this video, was very understandable for an electronics noob :)
Hi Dave,
48.9W is Power consumed by the BK Electronic Load, Regulator is producing 52.5W , Delta around 3W is the Joule effect produced in the cable , gone in heating the cables. on my point of view its not an error .
you doing a great job !
Ali.
Nice one Dave, thanks! A friend of mine uses a BEC that looks very much the same heatsink in his aeroplane. As his systems engineer :-) I had him try a couple different geometries to mount the BEC in a "prototyping fixture" I jury rigged. Best performance so far has heatsink fins getting air "scooped" & directed over fins
The heatsink data sheet calls for a 1" square heatsink, 25x25mm. Now I'm not confident of the scaling, but the heatsink on the device looks significantly larger than that to me - between 30 and 40 mm square.
Dave,
Castle Creations makes some of the better reliable BEC units in the RC Industry. I wouldn't worry about the lifespan of the caps. Most likely the failure of the BEC would be due to a crash. :-)
crashes are usually due to failures such as this. Most of us that have flown for years,rarely crash and usually its on landing due to cross wind landings. I havent had a crash in quite a while, the last one was caused by servo failure. I dont expect electronics to hold up for longer periods of time and generally most electric motor planes only fly 15 minutes tops while gas is 25-30 minutes at best. Like I said, most guys I fly with are pro's and generally dont crash. when you invest 3 to 5 k or more into a plane or heli,you usually dont go with cheap BEC's or U BEC's. I use An ESC and seperate BEC. I see too many planes not having enough air space to properly cool these types of BEC's.Happy flying!
I think it's safe to say BEC's are rarely run at their absolute max rating continously anyway. It's a varying load - I can't think of many situations where a heli or plane will be asking for maximum juice during an entire flight. Unless you're Alan Szabo.
Dave, thanks for looking at the Castle BEC Pro. I found it very useful indeed and confirms, for me at least, that Castle make good equipment.
3 amp is usually "higher spec" and safe from causing a reboot mid air for park flyers
normally four of these bad boys www.hobbyking.com/hobbyking/store/__38594__Turnigy_8482_TGY_113MG_DS_MG_Servo_2_2kg_0_10s_12g_AU_Warehouse_.html
bec feeds reciever and hence servos...unless its rebooting.. then it sings some crappy tune as your plane is in a death spiral...
Joey Vann you are absolutely right!
even so i flight helis and planes as well. but I never had the same curious though about all my BEC and speed controls like Dave is explained to us and I know that castle creation is one of the top brands out.
Was liking the look of your Flir thermal imaging cam Dave, until I found the price!!! £4000 in the UK!! I suppose it was a freeby then!
On using the voltage reg, presumably it would be mounted exposed to airflow created by prop wash? In which case, plenty of air flow!
Those output caps are Chemi-Con PS_ series (not sure of the last letter, series name couldn't fit on the top there) solid polymer electrolytics - i wouldn't be concerned with those kicking the bucket anytime soon. Besides, their tops are even "outside" so they get some cooling from the airflow.
I think the "continious" means 1 batteryload. most batteries only stand for about 10 minutes. when you change the battery its always reconommed to wait until the motor and speed controller are cooled down.
Would be interesting also to see how much noise this dc-dc converter produces at these high currents.
The caps were Nippon Chemicon PXE series.
I'm guessing the designers are assuming that this will never run at full load all the time. In an rc situation I think it would be hard to do that.
Those output current maxes also aren't likely at all output voltages. It's got an adjustable output voltage. It might output 15A just not set to 6.5V
All your calculations I believe were based upon 20* ambient temperature. In a hotter climate like 40* or maybe 45* ambient I just wonder if the converter will hold its regulation. Just a thought
also remember that its not always constant current in a rc plane or any rc really as servos dont always move and work hard constant
it is intended for an airplane so there is going to be a lot of airflow to cool it. it is also good that it has an overtemp shutdown.
I think it's safe to say if Dave built any kind of RC thing it would be overbuilt in so many ways and nearly impossible to kill. :P Overkill is my favorite kind of kill.
Heat sink size/weight will always be compromise since it will affect performance and flight time of planes, copters and quads.
That was very interesting! Dave, did you set your E8 to the correct epsilon factor? It can influence your temperature readingsby up to 10°C or more!
I think the ratings are nothing to do with the BEC output. A BEC is designed to run from the flight battery in an electric model that drives the motor, and provide a low voltage (typically 4.8v) to power the receiver and servos, which typically don't require a huge amount of current. Before BEC's came along, most people had to use a separate battery pack (often 4 x AA ni-cads) to power the receiver and servos. I doubt that a BEC would be expected to supply 8 amps or more !
I fly electric heli's up to 90 size (250+ amps into the motors). I use this type of controller all the time and they are normally used to power radio and servo's etc that create short term peak loads rather than continuous maximum loads. If you want to run them at full power continuously then yes you need to put them in air flow and they work fine. In other words they work fine for the intended purpose.
thanks for doing this test. In the electronics category you are among the 10% best on CZcams. thanks for sharing. i watch many of your videos.
Do take into account that this device is intended for use on an airplane/heli with plenty of airflow. If you're at max load and only going 5mph, something is really wrong. I think I would opt to operate it with the case off.
They are usually mounted inside the airframe, though, so 2~3 m/s is reasonable enough and it had the the added bonus of being tested without the airflow restriction of the plastic case inside a lab with air conditioning.
he adds a fan at 23:05
Ben The Shrubber He knows, he was arguing that fan wasn't good enough (hence the 5mph number), I was arguing it is, as in most cases the BEC is mounted inside the airframe with limited venting holes.
These are designed more for big helis running on 12s so on the higher input is wear they are most commonly used
ohh.. its just a bec... 8amp would be massive for servos... prob way overspecced product... or for big planes. Normally the BEC is with the speed controller, that would be more interesting viewing, seeing if it can send 120 amp continous.. some of the old boys like nicd batteries for BECs. The new coreless servos often with their wide voltage op ranges can run straight off the lipo.. with good torque characteristics
www.hobbyking.com/hobbyking/store/__20555__Turnigy_dlux_100A_SBEC_Brushless_Speed_Controller_w_Data_Logging_AUS_Warehouse_.html?strSearch=turnigy%20bec
get that up ya
A few of those watts will also disappear in the wires, so the heatsink would probably dissipate about 15 watts.
thanks Dave, for putting it through the tests that I can't do. sounds like it works OK, but needs lots of air to keep it cool!!!
Great testing there Dave. Never thought to stress test my BECs this way as I have the same PSU and a Maynuo EL (which looks to be a ripoff of your BK unit).
As many have said the Castle creations units are certainly one of the better brands out there for this sort of gear and they're not really expensive as you mentioned. I have a couple of those CC BECs but have never really stressed them - they are way over sized for even my large RC heli's. For reliability I've now removed the BECs and gone back to a dedicated control battery (2S Lipo). Some high performance servos and gyros ( for Heli's anyway) now directly run at the higher 7.4volts (as opposed to the traditional 5 volts) of a 2S battery so the BEC can be totally eliminated in my case. One less thing to go wrong IMO.
Once again - great video.
Dave hasn't been this excited since solar roadways
One minor thing... You should have put it inside it's original plastic case, thermal situation would be even worse for all the components inside.
Can't imagine the caps being the limiting factor for the device's lifetime. Would have thought its end would be more likely getting augured into the ground...
It would be most interesting to see the output on a scope under the various loads.
Big Clive would've shorted the output...
Assuming the temperature drop over the board I am actually surprised that it didn't burn.
Maybe the heat sink can be installed outside the fuselage ? Here you have a good air flow ;-)
Good video. Great insight. Notice the specs do not specify at what ambient temp.?
161C in freedom units is 321.8, for those wondering
72C -> 161F
*0:0 "Hoie"*
I'd watch it all, but I don't have 30 minutes spare for a CZcams video
watch at faster speed, then
1.25x or 1.5x speed is useful in these scenarios.
wah spoonfeed me free info the way I want it
Castle for the win! USA made right by me
Cool vid. One of the more interesting ones I've watched in a while :-)
Love the display on that B+K load. I have one of their bench multimeters (5491B) with a similar display, it's a joy to work with in a wide range of ambient lighting levels
Nice to see proper non goofed review of one of these... ;)
What about if it was running at 16 volts exactly instead of 15.2, would it manage to output 15 Amps or would the 0.8 Volts not be enough to cause a problem?
7:48 It's Nippon Chemicon PXE series Conductive Polymer Capacitor
If you could raise the output voltage from 6.5V to something more like 10V, it might very well be able to supply 15A. I wouldn't mind seeing more on this item if you find time :)
That thing could have used a synchronous rectifier FET, would have cut losses in half if not more, especially at high voltages deltas where the free-wheeling diode/FET carries the bulk of load current during the 80+% off-time.
The usual overstating of specs for most hobby products. What would help this survive is probably the fact that most of the load would come from the servos moving the control surfaces of the plane so only in short bursts as it maneuvers. Not much load when all the servos are neutral(eg flying straight).
I still don't understand why you would need that much power for a BEC. On my model car it was just a way to power the radio or other devices that would have minimal draw, eliminating the four "AA" batteries in series, unlike the motor which runs directly off the "Large main" battery or speed control. Other then the motor what other high drain devices would use this on a plane car or helicopter? At first I thought it was a speed control looking at the beefy wires.
On planes/helis you need a few servos and these draw a fair bit of current. Maybe on really big models that require strong servos.
Four high voltage standard servos producing 25kg-cm of torque and 0.05s/60deg will easily draw 2A during travel with no load on the output and much more peaks. Helis use four, and huge scale airplanes can use up to 8. That's some serious power. Cars are nothing compared to aircraft.
Awesome to see a test
Should we consider the heat generated from direct sun light or it has nothing to do with our case !
With model airplanes every gram counts so massive block of aluminium only shortens the battery life (which is 10-15 minutes anyway) or renders it unable to fly if it is too heavy.
Im Watching Your videos from centroamerica good job Dave
In the RC world that device is excellence.
I would like to know why there are two wires coming out with Servo ends?
The 160c was the hidden deice feature.
You disappoint me Dave - in Australia we call them AEROPLANES - not air-planes!
19:22 you forgot about power dissipated in wires.
(6.56V-6.11V)*8A=3.6W
efficiency is 79% :P
Anodized aluminum probably has a different emissivity than the default setting in your flir camera. Did you take that into account for your temperature measurement ?
How can you get your timing right? You said that the video would be about 30 minutes and it turns out to be! nice!!
8:28
Incorrect, polymer capacitors have lifetime ratings all over the place, just like electrolytics.
But yes, there are some large vent-less electrolytics out there. Specifically, the Sacon FZ series, which are known to go off like gunshots!
thank you for this test. very excellent