Running 12 cell limit ESCs with more cells by disabling BEC

[Finnster]

I saw this post over at RCZ in the brsuhless forum. Castle Creations has a factory answers thread and a ? was asked it it would be ok to run the MM on 14 cells if the BEC was disabled. The answer was:

Shawn Palmer
Member

Join Date: 07-16-2004
Posts: 48
RtsBasic:
It's out of spec, and not warranted, but yes - with the BEC disabled you can run 14cells. Just watch the overall rpm of the motor there (voltage x Kv).

Shawn

Link

Shaun is a CC rep. This has got me wondering if this is true for some of the other ESCs like the MGMs/Mtronics/ etc as they all have BECs and this has always proven to be the weak link. Ie ratings due to BEC and not controller max. The MGM120 and 160 is impressive but for the 12 cell limit. The CC controller for 14 cells at only $150 is an awesome deal. Right now the only options for the 14+ cell market is really the Q125 or the BKs Some more options would be very welcome.

[squeeforever]

Some would be ok, some won't. Originally the 9920's were 9918's. They infact were underrated and BK found that they could easily run 20 without risk. I don't doubt that you could run a little more than that to be honest. I'm sure there are others that can be run more than rated for. The main thing I would worry about would probably be the caps. All I can say is don't do it with a Schulze :p.

[BrianG]

Originally Posted by squeeforever

...The main thing I would worry about would probably be the caps...

And for that you can just look at the voltage rating on the cap itself. For safety and lengevity, I would only go as high as 80% of the rating. Of course there are other components in there that can blow before the caps do...

[MetalMan]

I'd be worried about the internal caps because you can't look up the voltage ratings on them too easily, and it wouldn't be easy to replace them if they blew. The only way to know if a controller can take higher voltage it is to try it out.

[Finnster]

From experience I know the 9920s have 50V caps. My Mtroniks has a "25/085/21 1430"

Not sure what that is, but I would guess a 25V 85C cap.

I would be surprized too that you could not push the voltage a bit. Its not like they would produce a esc righton the edge of specification (at least oyu hope not.) I don't know if you could push 16 or 18 cells, but I do wonder if alot of these esc are built with very simular components and its really the becs that are holding them back. The rec'd here are not to run a Q125 with the BEC w/ high cell counts, yet the esc is listed as such. Who's to say that Quark is just pushing the specs of their esc (knowing bec will prob get disabled) while MGM is not. The fact that it needs an extrenal HS to be reliable suggests they are pushing the design spec to begin with. Not saying its a bad controller, but a $280 esc that still needs a ubec and extra HSF to work as advertized I think clearly demonstrates they are not being conservative with the ratings.

[squeeforever]

Originally Posted by Finnster

From experience I know the 9920s have 50V caps. My Mtroniks has a "25/085/21 1430"

Not sure what that is, but I would guess a 25V 85C cap.

I would be surprized too that you could not push the voltage a bit. Its not like they would produce a esc righton the edge of specification (at least oyu hope not.) I don't know if you could push 16 or 18 cells, but I do wonder if alot of these esc are built with very simular components and its really the becs that are holding them back. The rec'd here are not to run a Q125 with the BEC w/ high cell counts, yet the esc is listed as such. Who's to say that Quark is just pushing the specs of their esc (knowing bec will prob get disabled) while MGM is not. The fact that it needs an extrenal HS to be reliable suggests they are pushing the design spec to begin with. Not saying its a bad controller, but a $280 esc that still needs a ubec and extra HSF to work as advertized I think clearly demonstrates they are not being conservative with the ratings.

On the contrary, the 9920 for instance are only rated for 12 cells on the BEC. A external BEC is required for higher. Let me explain why. The total voltage of the batteries has to go through the BEC or UBEC either way. With a internal BEC, the ESC has a fairly small BEC internally so as to maintain a small ESC, were as obviously, the smaller internal BEC is also gonna have problems disipating the heat. Heat comes to it in the form of watts. Which is gonna be killer on it with a large spike. At least this is how I understand it. Alot of this is still over my head.

[Finnster]

Originally Posted by squeeforever

On the contrary, the 9920 for instance are only rated for 12 cells on the BEC. A external BEC is required for higher. Let me explain why. The total voltage of the batteries has to go through the BEC or UBEC either way. With a internal BEC, the ESC has a fairly small BEC internally so as to maintain a small ESC, were as obviously, the smaller internal BEC is also gonna have problems disipating the heat. Heat comes to it in the form of watts. Which is gonna be killer on it with a large spike. At least this is how I understand it. Alot of this is still over my head.

Sorry, I was trying to make that part of my point, just didn't do it well. I did see where the 9920 is listed here as only upto 12 cells with the bec.

My thinking is that the 9920 does have a BEC jumper, so an easily disabled BEC. The then advertize it can run upto 18, then 20, w/o the BEC. The fact that ~every non-BK ESC is only rated upto 12 cells seems to me not necc a coincidence. Most of all these other ESCs are a bit more "plug and play" than the 9920 (or maybe better said a more "mainstream commercialized" unit) as they are encased, no jumpers and such. Perhaps due to marketing/warranty decisions these other makers are not advert higher cell counts w/o a mainstream acceptable method of disabling the BEC, and don't want to face lots of warr claims/ bad press b/c of kids running 18cells on the BEC and frying the ESC when it was advert to due so.

I'm wondering if Quark just decided to do this where others didn't. :032: Looking on their website, no where does it say you have to disable the BEC to run 18 cells. Perhaps this was a clever marketing strategy, figuring most users would be savvy enough to figure out the BEC, but still have the only other big name esc that can run 18cells other than the BKs. Or maybe not. I have plenty of time to think about it since I don't have a esc to run currently....

Originally Posted by Finnster

I'm wondering if Quark just decided to do this where others didn't. :032: Looking on their website, no where does it say you have to disable the BEC to run 18 cells. Perhaps this was a clever marketing strategy, figuring most users would be savvy enough to figure out the BEC, but still have the only other big name esc that can run 18cells other than the BKs. Or maybe not. I have plenty of time to think about it since I don't have a esc to run currently....

in the quark manual it says that you can not run the bec with over 12 cells or 4s lipo.

[squeeforever]

The fact that ~every non-BK ESC is only rated upto 12 cells seems to me not necc a coincidence

It isn't a coincidence. The use the same kinds of internal BEC's (can't really remember if there a Linear or not) and that type they use in all the different ESC's can only handle so much heat before they go poof.

[Finnster]

Originally Posted by cemetery gates

in the quark manual it says that you can not run the bec with over 12 cells or 4s lipo.

;) .


Only way to know is try it I guess. Does mike sell just the CC ESCs?

[BrianG]

Originally Posted by squeeforever

It isn't a coincidence. The use the same kinds of internal BEC's (can't really remember if there a Linear or not) and that type they use in all the different ESC's can only handle so much heat before they go poof.

Yeah, most I've seen are linear.

The more cells you use, the more power is wasted by heat on the BEC. The power dissipated on the BEC is (v_batt - v_bec) X bec_current. v_bec is usually around 5-6v. So, if you are using 12 cells (~14.4v) and your servos are drawing an average of 2A, that is almost 17 watts the heatsink has to dissipate. If you only use 8 cells (~9.6v) with the same 2A servo draw, that's only a little over 7 watts. Since the ESC heatsink is usually pretty small, it needs all the available surface area just for the motor controller. Any added power is just going to make it heat up faster/more. If you can't afford or don't want to use a digital UBEC, at least hook up a simple outboard linear regulator.

The internal surface-mounted caps are usually not electrolytic types and aren't quite as sensitive to voltage. But like MetalMan said, the external caps can't really be used to judge the ESC voltage capacity. Maybe the manufacturer got a really good deal on a bunch of 50v caps and decided to use those instead of the "proper" 35v or 25v caps. Also, all those FETs have a max working voltage as well, and unless you can read the part number and get a datasheet for them, there is no way to really tell.

But if you want to experiment, and have deep pockets, then more power to you! :)

[glassdoctor]

I just got a replacement Mtroniks Truck and I think it might have smoked due to a BEC overload. Mtroniks made the comment that it was possible and that I might want to use an external power source.

They said the BEC can heat up pretty quick and that since I was likely pushing the limits of the ESC current handling, that the hot controller combined with the high BEC loads (4s lipo and a powerful digital servo) possibly explained the controller's meltdown.

I am going to try a rx pack/ external bec... I hope that fixes it...

[Finnster]

Originally Posted by BrianG

Yeah, most I've seen are linear.

The more cells you use, the more power is wasted by heat on the BEC. The power dissipated on the BEC is (v_batt - v_bec) X bec_current. v_bec is usually around 5-6v. So, if you are using 12 cells (~14.4v) and your servos are drawing an average of 2A, that is almost 17 watts the heatsink has to dissipate. If you only use 8 cells (~9.6v) with the same 2A servo draw, that's only a little over 7 watts. Since the ESC heatsink is usually pretty small, it needs all the available surface area just for the motor controller. Any added power is just going to make it heat up faster/more. If you can't afford or don't want to use a digital UBEC, at least hook up a simple outboard linear regulator.

The internal surface-mounted caps are usually not electrolytic types and aren't quite as sensitive to voltage. But like MetalMan said, the external caps can't really be used to judge the ESC voltage capacity. Maybe the manufacturer got a really good deal on a bunch of 50v caps and decided to use those instead of the "proper" 35v or 25v caps. Also, all those FETs have a max working voltage as well, and unless you can read the part number and get a datasheet for them, there is no way to really tell.

But if you want to experiment, and have deep pockets, then more power to you! :)

Right, so that's why it seems the BEC is the limiting factor, not the other components (to a reasonable level.) Remove that limiter, and you can push a bit higher. I'm not suggesting that the ESC is merely rated by their caps, else BKs would be rated as a 50V esc.

By disabling the BEC you are removing alot of extra heat, I wouldn't be surprized if it could handle an extra 2.4V (that's all we are really talking here.) To the BEC, that is several more watts (5 or more), but the rest of the system would prob be within specs for that small amount of voltage (its not like they are using 15V caps.) Considering it is coming from a CC rep/engineer (he's mentioned in the CC thread on this board) this is prob not pie in the sky. The CC ESC is rated upto 100A, it would seem capable of running a 8xl or 9xl motor on 14 cells if done with care.

I had a GTB/5.5 that I ran on 7 cells fine, tho Novak only rates them for 6 cells. ;) It would not make sense for a reputable dealer to sell a product that was right on the edge of design capacity. You build in a safe margin that is reasonable as budgetary constraints allow. It would be an interesting experiment, and if it worked well, it would be a very sweet controller for a mild setup.

[GriffinRU]

Well you covered capacitors and BEC very well, couple things to add FET's ratings and FET's control logic capability to drive high side of the bridge. ESC rated at 6-7 cells not designed to run more then that without commutating problems. Then we have ESC with good and cheap FET drivers. Good automatically adjusts to required voltage, only caps and FET are the limits. And cheap with fixed voltage which is good only up to given maximum number of cells.
And do not forget about electrical noise, higher voltage requires better suppression.
So you can try, but keep that in mind.

Artur

[captain harlock]

Does this mean I can run 6s lipos in my 149.18 Schulze controller?
It is rated for 5 lipo cells, though. can an extra cell destroy it?