RunRyder RC
 14  Topic Subscribe
WATCH
 5 pages [ <<    <     1      2     ( 3 )     4      5     NEXT    >> ] 5716 views POST REPLY
Helicopter
e-
Align
Other › CC phx guys, I need you to look at something.
03-29-2005 12:47 PM  13 years agoPost 41
dkshema

rrMaster

Cedar Rapids, IA

My Posts: All  Forum  Topic

Here is a reply in its entirety from Castle Creations regarding their voltage regulators on the CC 35 ESC :
David,

I'd like to address each one of your concerns about the BEC.
First of all, the 3A rating comes from the fact that the National LM2940IMP-5.0 actually doesn't start to current limit until above 1.6A, so the surge capability (what really counts in a BEC) is conservatively rated at 1.5A per regulator.
Secondly, you state that the LM2940s are obviously not well heat sinked. Well, it may not be obvious, but there is 2.4 square inches of 2 oz copper heatsinking the regulators (the control board is 4 layer 2oz/layer.) If you read the National datasheet, you would see that 2.4 square inches of copper is more than adequate heat sinking. In fact, in our testing on the original design of the Phx-35/45 control board, we found that the BEC could easily dissipate more than 5 watts of waste heat (in fact, we had to push it over 10 watts continuous just to start seeing regulator shutdowns.)

The two LM2940s are indeed paralleled. However, your assumption that paralleling LDOs is a bad practice and can cause problems is simply not correct for this type of application. Some applications, where the continuous current is higher than the amount a single regulator can supply might have problems with paralleled regulators, and in those cases load sharing resistors are required between the regulators. In the case of the regulators on the 35/45, the load imbalance between regulators is desireable -- one regulator will take the lion's share of the load, with the second kicking in to supply extra current if the first regulator starts to current limit. Because the continuous current load on the regulators is below the normal output current of a single regulator, there is no issue with paralleling.

Remember, a BEC is required to supply very high current surges at low duty cycle, with low current supply at high duty cycles. Because of this requirement, the BEC design on the 35/45 is superb at what is is required to do.
BTW, your assumption that a pass transistor would be fine is incorrect. A pass transistor can work, but is a really bad idea because there is no on-die temperature limiting, which makes COMPLETE failure a possibility. Just ask all the people that lost aircraft to early Jeti brushless controller designs which used a pass PNP transistor and a 100ma regulator to drive the PNP. The Jeti controller had little heat sinking for the PNP, and when it overheated it failed, causing a complete loss of control of the model.
Thanx!

Patrick del Castillo
Castle Creations
402 E Pendleton Ave
Wellsville, KS 66092
(785)-883-4519 (v)
(785)-883-4571 (f)

Richard Hofer wrote:

>
>
> ----- Original Message -----
> *From:* shawnp
> *To:* richard@castlecreations.com
>
> *Sent:* Monday, March 28, 2005 11:08 AM
> *Subject:* Fw: Phoenix 35 controller BEC rating]
>
> Will you talk tech with this guy please in a reply - I don't want
> to get in-depth without the verbage and experience behind me to
> back it up.
> Shawn
>
I reserve comment on the above.

Dave

PM  EMAIL  GALLERY  Attn:RR  Quote
03-29-2005 01:18 PM  13 years agoPost 42
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

Secondly, you state that the LM2940s are obviously not well heat sinked. Well, it may not be obvious, but there is 2.4 square inches of 2 oz copper heatsinking the regulators (the control board is 4 layer 2oz/layer.)
BS to that, if that was true the ESC would way over 1/2 a pound, 4 x 2oz = 8oz.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
03-29-2005 01:38 PM  13 years agoPost 43
MMarshall

rrNovice

Herndon, VA

My Posts: All  Forum  Topic

I believe the reference to the 2 oz copper heatsinking is not to mean literally that it weighs 2 ounces but maybe a measure of its thickness?

Just for the record I run a P35 with 4 HS 55's and a 240 Gyro with no problems, I also ran a P25 with 4 HS 55's and a 240 with some occasional glitching that may or may not have been a BEC issue but when I upgraded that Heli to HS 56's I added a UBEC to be safe.

I like everyone with expensive equipment worry about a failure and have to rely on the makers limitation ratings, I think CC has kept everyone informed on the limitations of thier equipment and have even looked at improving some to meet growing needs. Since my return to RC with electrics I have run both Jeti and CC esc's and have never had a failure with either, in fact I have only seen 1 partial failure of a P10 when the user hooked the ESC up to a charger and that only burnt some of the chips... the esc still functions albeit missing some throttle steps.

Just MHO

Gravity is a harsh mistress

PM  EMAIL  Attn:RR  Quote
03-29-2005 01:53 PM  13 years agoPost 44
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

First of all, the 3A rating comes from the fact that the National LM2940IMP-5.0 actually doesn't start to current limit until above 1.6A, so the surge capability (what really counts in a BEC) is conservatively rated at 1.5A per regul
??? Let's see 11-12V in and 5V out at 200ma, so no it does not limit
current, it's a VOLTAGE regulator.

And I dare anyone to put a 5 ohm 5W resistor on the ouput of the bec, I mean that would be a very conservative 1A / 5watt load divided by
2 (since they share the load between 2 IC's) and then hold your
finger on the BEC.

The temp will within 5 secs go over 250F , WAY over the chips allowed max temp of 185F.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
03-29-2005 04:30 PM  13 years agoPost 45
Inspector Fuzz

rrKey Veteran

NLA

My Posts: All  Forum  Topic

2 oz copper
This refers to the weight of 1 SQARE FOOT of the material.. I was considering a 20 oz copper roof for my house and ended up going with 24 guage steel, due to cost..
JEFF

PM  EMAIL  Attn:RR  Quote
03-29-2005 04:46 PM  13 years agoPost 46
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

If you read the National datasheet, you would see that 2.4 square inches of copper is more than adequate heat sinking. In fact, in our testing on the original design of the Phx-35/45 control board, we found that the BEC could easily dissipate more than 5 watts of waste heat (in fact, we had to push it over 10 watts continuous just to start seeing regulator shutdowns.)
This I HAVE to see as my tests show a different story. And I don't believe you can count the copper in the other layers of a multi layer board so there is NO 2.4 sq inches for the BEC, in fact if you look at the total area for the BEC I'd say it's no more than 1 at max. You can't count the total area of the board if it's not soldered to or heat sinked to it. Air transfer does not count.

So please , someone try my test and tell me your findings on temp
of the bec.

And to answer some, will it last as is ? Maybe , maybe not, it's like
a nascar engine running 10K, will it last the race , you hope.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
03-30-2005 02:48 AM  13 years agoPost 47
lbnaz

rrNovice

Corning. NY

My Posts: All  Forum  Topic

This person looks like he has done a similar test, it is a discussion on BEC & UBEC with the Castle ESC. The results are the same & the recomendation is to run an external BEC. I run a UBEC for this reason also.
http://www.rcgroups.com/forums/show...1e6646&t=348938

PM  EMAIL  Attn:RR  Quote
03-30-2005 04:33 AM  13 years agoPost 48
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

Great, So I'm not the only one. I'm sorry but their engineers are wrong, how can they not read the same things on the same info sheet for the same IC ? I mean it states operating range to 185F , they say only after 400F when the solder melts, WHO the &*^# are they kidding? Not me.

It's a 1Amp REG, where do they get 2.5A ?
I talked to Bernie at CC today about this. He said that there are two BECs on the 35/45 (which was already pointed out above - just repeating the conversation), and that each one was rated at 2.5amps. When one gets too hot and shuts down the other kicks in (actually, he said both are online all the time, but since they are not perfectly matched one of them works first until it shuts down, then the other takes over until the first one cools).I talked to Bernie at CC today about this. He said that there are two BECs on the 35/45 (which was already pointed out above - just repeating the conversation), and that each one was rated at 2.5amps. When one gets too hot and shuts down the other kicks in (actually, he said both are online all the time, but since they are not perfectly matched one of them works first until it shuts down, then the other takes over until the first one cools).

I believe what is actually happening, based on aragon's research indicating a 1 amp max per regulator, is that one BEC FET is active (again, since they are not matched perfectly) up to the 1 amp range until it gets too hot, then the other kicks in, supporting Bernie's claims. When the first one cools back down it takes back over. So, I believe the 2.5 amp rating is for both BEC FETs combined, but that is not how they operate - only one is operating at capacity at a time, so the 1 amp limit is the "real" continuous limit.

He did indicate that these things do get very hot, and that they can get to 400 degress (ouch) before there is circuit damage (basically the solder starts to get soft and melt, then the sparks fly). He also said that cooling helps quite a bit.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
03-30-2005 05:12 AM  13 years agoPost 49
WRXRonald

rrKey Veteran

Auburn, OH

My Posts: All  Forum  Topic

alright...
well i am beginning to believe there is a problem out there.........

So what would be the results of failure during flight...

(i've got a bad feeling i'm ALREADY starting to experience these probs)

twice now in flight since i put the new ESC in..... my bird has just VERY STRANGELY stopped doing what i was telling it to do...

once it fell from high in the sky, i had no clue what happened or why it plunged... basically i was just hovering at about 75 feet up or so, gave it a bit of input and from there it just tailed over backwards, did full flip in the air and somehow i don't know if i righted it or it righted itself but by the time it got to the ground in landed pretty hard into soft/mushy ground on the landing gear.... only snapped one landing gear and the vertical stabilizer... at the time i chalked it to human error, not quite understanding what exactly i had done wrong-- i AM proficient at hovering, slow forward/rearward flight, nose in, some small slow funneling, etc...... it didn't seem to me that i had done anything wrong. Could this be loss of the BEC and therefore loss of the controls?


Then tonight, i'm outside just doing very basic hovering in the dark (lights on in my driveway)..... again, all of a sudden i give it an input and it never came back from the input, just kept going hard sideways, i again somehow got very lucky to land it on the gear, AGAIN breaking another landing gear set.... and this was quite clearly lack of the stick inputs i was giving reaching the heli... i KNOW it was not tx/rx range issues, (i was about 4 feet away from the bird and about 5 ft off ground in a hover).... again, could this be the BEC cutting out... i have it tucked away inside the frame which is gonna be a bad mistake apparently, i'll get it out of there now and under the blades for better cooling...... but i just hate to mount anything outside the frame, i usually just separate my Rx/ESC w/alum foil, never had any glitching probs so far...

to fix the issue once and for all--- a separate BEC? where should i get one? how much will it cost? which one should i get? how much weight is it going to add?

thanks.

good research rckrzy1 i think you're more on the mark than most would like to admit.... i'm getting a bad feeling that my control probs in the last few days are related to this issue, i don't know how else to explain them.....

if the BEC cut out, would it be fried for good? or would it start working again a few moments later after a quick cool down? cuz everthing seems to work ok after the heli sits for awhile and i take it back out the next day...

PM  EMAIL  GALLERY  Attn:RR  Quote
03-30-2005 05:15 AM  13 years agoPost 50
dkshema

rrMaster

Cedar Rapids, IA

My Posts: All  Forum  Topic

I've been holding back since I got the reply from CC with respect to the BECs.

You can run linear monolithic regulators in parallel -- but there are problems with doing so.

The first is that one will tend to supply most of the current while the other one idles. This is due to the differences in the individual device reference voltages and differences in manufacturing from lot to lot.

As CC stated, one will take the brunt of the current, till it goes into thermal shutdown, at which point the other begins to source the current. The first one will cool down to a point where it begins to be the primary source again. It will again heat up to the thermal shut down point and the cycle will repeat.

A design that relies upon a device's built-in thermal shutdown circuits is a design as standard practice that I would personally avoid in my profession. Generally, a sound practice would run the devices at a de-rated level and avoid regularly going into thermal shutdown.

Since one regulator is cycling on and off, and expecting the other to take up the slack, the overall regulation suffers -- the output voltage of both regulators will not necessarily be the same due to manufacturing and component tolerances. The power to the system will fluctuate.

If one really wanted to connect two linear regulators in parallel, a small series resistor is inserted in series with each regulator's output and the load.. The resistor forces the two regulators to share the load. This would keep one from cycling off and on due to overcurrent related thermal shutdown. The down side of the small resistor is that there is a voltage drop across each resistor, and the voltage to the load is reduced by that drop. The drop varies linearly with current. The stability of the system is also compromised.

As stated in the National Semiconductor brief, a short circuit on the output can lead to device failure as the control circuits in each regulator begin to interact. The result might be uncontrolled oscillation in the control circuits and device failure.

I guess I could go into a discussion about the data sheets, the thermal curves, heatsinking and such, but that would get a bit tedious and long.

In practice, so far my two CC35s seem to work (well, at least I don't have BEC problems, I do have SW problems with regard to shutdown for controller current limiting and my 400DH motor....). The regulator design is one that I would not choose in my own line of work (nor would it make it through any serious design review, let alone reliability and failure analysis). There are serious engineering papers out on the web that discuss parallel linear regulators. I don't think any of them recommend a simple parallel connection as a common practice.


Dave

PM  EMAIL  GALLERY  Attn:RR  Quote
03-30-2005 02:33 PM  13 years agoPost 51
MMarshall

rrNovice

Herndon, VA

My Posts: All  Forum  Topic

I tested my P35 last night with 4 HS55's and a GY 240 locked inputs over to one side with motor off and Esc gets warm, not hot just warm didnt think to temp it but its way different then my P25 was, that got plain hot doing same test thats why I went with a UBEC with it.

Gravity is a harsh mistress

PM  EMAIL  Attn:RR  Quote
03-30-2005 03:37 PM  13 years agoPost 52
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

I took mine apart and added heat sink compound to the BEC on the back side of the controller and then cut a piece of copper to add to the top of the controller, that was easy as the hole in the heat shrink from the hot bec made it a direct contact to it without any chance of shorting
other components and again used heat sink compound there to.

Any CPU overclocker can attest, heat sinks without proper heat transfer compound is not good.

BTW since that email from Patrick is written by the owner and lead
engineer I doubt we will get anywhere. As a technician I know better than to try and change an engineers view point. Hell we techs only know how to fix engineers mistakes, right?



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
04-21-2005 07:21 PM  13 years agoPost 53
electrotech

rrNovice

Allen, Tx - USA

My Posts: All  Forum  Topic

I use CC25 with Trex V1, TP 3s 2100, Align400, no problems
I currently run a CC25, as this was recomended by a local shop. I wanted to use a higher current rated ESC for margin, but have been sucessful with the 25A for at least 40 full flights (hover and FF, no 3D since i'm a newbie).

Anyhow, I recently changed the ESC settings from fixed throttle to governor high, and noticed the batteries, motor, and ESC are running extremly hot when compared to Fixed mode (which barely became warm). Only other change was servos from HS55 to HS56HB (which draw more current) to fit the Rotorworkz CF frame.

Tonight I will change the ESC back to the Fixed mode and see if there is a difference since I melted the heatshrink in governor mode (won't be using this setting agian). If this does not help, I will bypass the BEC and add the receiver battery (hate to add that much weight). This should tell me if there the problem is in the governor or BEC. Though, since the motor is running much hotter, I expect the governor mode is eating the additional current by forcing timing of the motor, probably a "backwards EMF" problem.

Will Work for Heli Upgrades

PM  EMAIL  HOMEPAGE  Attn:RR  Quote
04-29-2005 05:17 PM  12 years agoPost 54
electrotech

rrNovice

Allen, Tx - USA

My Posts: All  Forum  Topic

I use CC25 with Trex V1, TP 3s 2100, Align400, no problems
Changed settings from govoner mode to fixed mode, no changes, motor and BEC still hot. Bought a CC35 and running fixed mode, Motor hot, BEC barely warm. HS56 servos probably using more current then the CC25 could handle.

Will Work for Heli Upgrades

PM  EMAIL  HOMEPAGE  Attn:RR  Quote
04-29-2005 05:24 PM  12 years agoPost 55
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

Changed settings from govoner mode to fixed mode, no changes, motor and BEC still hot. Bought a CC35 and running fixed mode, Motor hot, BEC barely warm. HS56 servos probably using more current then the CC25 could handle.
Define barely warm ?

BOTH my CC35's with idle current only (170-200ma) And NO motor attached at all makes the BEC run 170F , and with a chips who's max operating temp is 185F , that's just plane poor engineering.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
04-29-2005 05:55 PM  12 years agoPost 56
electrotech

rrNovice

Allen, Tx - USA

My Posts: All  Forum  Topic

CC35
What is location and orientation of ESC/BEC.
I have min mounted micro side up (towards motor) undernearh chassis to alloe heat escape.

or Maybe your CC35 is defective?

when I first received mine, one fo the FETs were reworked, one pin not connected, so I had to remove the heatshrink and since the heavy copper use with no thermal reliefs to the part, I had to reflow this FET to get it properly soldered. So defects are not out of the question, just a matter of luck and PPM failures acceptable by CC.

Will Work for Heli Upgrades

PM  EMAIL  HOMEPAGE  Attn:RR  Quote
04-29-2005 06:46 PM  12 years agoPost 57
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

What is location and orientation of ESC/BEC.
I have min mounted micro side up (towards motor) undernearh chassis to alloe heat escape.

or Maybe your CC35 is defective?

when I first received mine, one fo the FETs were reworked, one pin not connected, so I had to remove the heatshrink and since the heavy copper use with no thermal reliefs to the part, I had to reflow this FET to get it properly soldered. So defects are not out of the question, just a matter of luck and PPM failures acceptable by CC
Mine laying under the main gear / fan in a vertical fashion.

I have 2 cc35's with the exact same readings.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
04-29-2005 09:27 PM  12 years agoPost 58
electrotech

rrNovice

Allen, Tx - USA

My Posts: All  Forum  Topic

rckrzy1,

heck, you're in hurst, i'm in Allen, maybe we can fly someday.
Course it windy as chit today, last night was nice though, practicing some figure 8s and FF, a real novice, but at least i'm no longer hovering in the Garage.

Anyhow, you using a pyrometer to measure this or thermal couple?
Orientation seems like it should be fine. I'm thinking you have too much demand on the BEC, maybe a bad servo drawing the current.
Which servos you running?, do you ever get servo glitches. This occured on my CC25 when overheated.................

Will Work for Heli Upgrades

PM  EMAIL  HOMEPAGE  Attn:RR  Quote
04-29-2005 09:47 PM  12 years agoPost 59
rckrzy1

rrElite Veteran

Hurst Texas

My Posts: All  Forum  Topic

Well without re typing everything, I used a variable load and power supply and measure the total current to the BEC with no motor at all.
And the BEC as stated by CC can handle 2.5A , well that's pure BS.
The BEC's are listed by the manufacture as 1Amp, and putting 2 in parelell does not give you 2.5 either. And any load over 250ma exceeds the max temp of the bec. I used a infrared / laser pointed calibrated temp gun from work. If you look back you will see applying only half of what CC says will work over heats the BEC immediatley.

And CC told me it was a 3amp BEC, BS to their EE who said that.
At that rate the BEC would fail before it got off the ground.

I mean read how many people are melting the heat shrink off their BEC.
It's under engineered because they assumed the loads would average less than they currently are. I mean who rates a device saying it can handle 3 or 4 servos, that's way to broad of a statement. And the only this CC35 would handle the 3AMPS they claim would be to use huge heat sinks with fans. You simply can't series regulate like they claim.



Wildcat Fuels

PM  EMAIL  GALLERY  Attn:RR  Quote
05-11-2005 06:55 PM  12 years agoPost 60
electrotech

rrNovice

Allen, Tx - USA

My Posts: All  Forum  Topic

rckrzy1

You ET or EE?
I'm ET and PCB designer.

I had the CC25 HS melt with the HS56 servos, but noe with the HS55 servos, this must be due to different current ratings of those servos, which would account for the Torque delta. Internally the motor is much larger in the HS56.

I've been using the HS56 with a CC35, no problems yet and no servo glitching due t BEC overheating as I had with the CC25 and HS56s.
This BEC is not even getting warm.

I will check the BEC part numbers to see if this release is any different than yours.

Maybe CC rune the 25 and 35 on the same line and accidentially placed the wrong component during assembly. Anyhow, maybe you can simply replace the BEC supply with a beefier chip, then you cna get use of these controllers.

Will Work for Heli Upgrades

PM  EMAIL  HOMEPAGE  Attn:RR  Quote
WATCH
 5 pages [ <<    <     1      2     ( 3 )     4      5     NEXT    >> ] 5716 views POST REPLY
Helicopter
e-
Align
Other › CC phx guys, I need you to look at something.
 Print TOPIC  Make Suggestion 

 14  Topic Subscribe

Saturday, April 21 - 7:08 am - Copyright © 2000-2018 RunRyder   EMAILEnable Cookies

Login Here
 New Subscriptions 
 Buddies Online