Lets talk Servos

[Tekin Prez]

There really is no good reason we should need seperate BEC modules for these rigs. A properly designed servo should be able to operate with the standard BEC of a speedo.

So what are the best current servos and why?

How fast do they need to be?

Lets figure something out that is right for these applications and makes everything simplier and more robust.

Tekin Prez

[binaryterror]

I heard a rumor Tekin was working on the servo market...odd

Personally anything made by JR has shown me the best speed/strength/ability. I prefer their entire high end digital line 8800T/S, 9000T/S, 9100T/S and the phenomenal 8711.

I have had quite a few Hitecs...all failed at some point. Their service was great, but since I was racing I couldn't support the back ups needed to guarantee finishes.

What I'd like to see is an aluminum front piece where the servo mounts and for better supporting the output shaft...this would help out immensely.

Specs I feel for a 2.2 Comp:
.15 or so Speed @ 6v
300 or more Torque @ 6v

Maybe just maybe make it handle a full 7.2v or better 8.4v for lipo.

[Scottmisfits]

I am running the Futaba 9351 in my 2.2 and am not lacking at all. It's rated slightly under 200in/oz's and a 0.12 for speed at 6v. The servo's can handle more voltage, it's some of the receivers that can't.

And a properly designed servo can't hack it at a 1 amp input when it draws upwards of 3.5 amps.

[saul]

well..in rock crawling...servos are not just for steering
they are for shaking a truck off a rock
sliding a truck away from a rock when u get pined
and for steering

the stronger and faster the better

[Tekin Prez]

Originally Posted by binaryterror

I heard a rumor Tekin was working on the servo market...odd

Not really when you consider that a servo is in fact a reversing speedo with position feedback and a brushed motor. Also FYI I am actually a Mech engineer although I function as a EE nowdays.

They both use a PWM drive and a standard H bridge configuration with the fets. Servo circuits are typically a lot less intelligent in many respects. We call them gear grinders since they only go full speed left or right and are not smart enough to slow down or brake. When holding position they are bouncing back and forth between full speed left and right. Imagine if you drove your car that way using full fwd and full reverse over and over many times per second... How fast would the motor and speedo get hot from the huge current spikes, basically locked rotor amps over and over.

The other major factor is the low quality of many gear trains. There are a lot of gear meshes in a servo and most are pressed, low quality gears. A good machined gear set can make a big difference. The gear reduction in most servos is around 300 or 400 to 1. The Hitech 625 and 645 are the same servo with different gear reductions. Speed vs torque.

High speed equals high gearing or a low turn servo motor. High speed and high torque equals lots of power. If you could tolerate a slower transit speed, you can have all the torque you want and draw a lot less power. If you have to have the fast speeds to wiggle on the rocks it may be tougher to reduce the current draws and maintain the torque.

Tekin Prez

[Tekin Prez]

Originally Posted by Scottmisfits

The servo's can handle more voltage, it's some of the receivers that can't.

Not typically true. Most servos have dual package fets instead of 4 seperate fets for the H bridge. They tend to be low voltage fets and do not handle more than 6v very well or very long. The receivers is another area of concern. They can typically handle the higher voltage, but cannot pass the higher currents without dropping the voltage within themselves due to internal resistance.

That is another option however is to use high V fets and connect the servo power directly to the battery and skip the BEC module and speedo BEC. The old KO fet boost servos were built to handle this. It is really a cost issue since most products are built to be cost effective.

Less speed is the easiest way to reduce power and still have all the torque.

For now we are talking about what is the right answer and designing without comprimise for performance or concern for cost. Then we decide if we can justify it. The speedo BEC needs to actually supply more power to the servo which can be expensive and large or we need to make better use of the power we can provide.

Tekin Prez

[SVEN]

Originally Posted by teamtekin

If you could tolerate a slower transit speed, you can have all the torque you want and draw a lot less power. If you have to have the fast speeds to wiggle on the rocks it may be tougher to reduce the current draws and maintain the torque.

Tekin Prez

I'm willing to sacrifice some speed for more torque, for a Super Class rig at least. 2.2's won't benefit from more torque as much as the Super's would, due to smaller size.
I run two Hitec 5995TG servos on 7.2 volts (through a CC BEC) in my super and have stalled out the servos on occasion. If I can't get my tires to turn when I want them to, it doesn't really matter how fast the servos can turn.

[mysharona]

I would LOVE to see a super high torque (400+) oz/in, with an aluminum case/mounting ears, and a tough gear train for under $100. personally, for heavy rigs like supers, I could care less if the transit time was around .3-.4 secs, gimmy a servo that is geared way down and a high turn motor any day!


since you seem to be the servo electronics buff, is it possible to say take a hs645 and rewind the motor for more torque?

[mysharona]

Originally Posted by SVEN

I'm willing to sacrifice some speed for more torque . If I can't get my tires to turn when I want them to, it doesn't really matter how fast the servos can turn.

Exactly my point! Bring on the slow torquey servos!

cough cough hint hint a company could really capitalize on the crawling servo if they were to produce somthin' like this! cough

[Rckcrwlr]

First off it is great to have another vendor on the board...this is great.

I am running 5955's on the super and smoked several of them last year. The key to performance of these is to get the heat away from the fets. a backplate heatsync built in would be great...possibly even an integrated fan like an esc has would work to. On those hot days when there is no wind...that would be great.

I like the idea to build for purpose and then cost...I have 5 of the 5955's that's a lot of $$$ tied up... Build it to perform flawlessly and I will pay the money...

Thanks

[saul]

oh...and as cool as it is to have a black servo....

we crawl on rocks ...in the sun and in the middle of summer

oh..and unlike race trucks....our servos are not covered by a body most of the time

so the sun basicly cooks our black servos

i know jr makes red cases...but why not white? or silver to reflect the sun and not just absorb the heat .. making our servos less efficent?

[JohnRobHolmes]

Just saw this, very good thread.

The best servos are the most powerful right now (with efficient gearmesh). 5955's, 8711, etc. Anything upwards of 300oz/in is good and more is better. Separate BEC's are needed because of the stalling amperage just one powerful servo has, the 5995 will stall 7 amps plus at 6v. Some guys are running them higher.


I would love to see a servo with a .5 second unloaded transit time @ 6v. We don't need anything faster, and some guys may even want slower. Most servos won't turn faster because of the load anyway. Working on voltages up to 15v would be very nice as well, so that 3 cell lipo would be safe. It would be easy to power from the main pack, and it would get rid of the pesky BEC issues.

[montereycrawler]

The only time I like the servo to be kinda fast on my 2.2 is when I wiggle or shuffle the front. The 8711 at this point seems to have the best of both worlds. I run it at 6.2v and it is pretty fast for having so much torque.
Even on my 2.2 though I could use more torque.

I think there is a market for a high end (expensive) servo if it sets itself apart from the others enough to justify the expense. Like JRH said, eliminating the BEC and being able to use it with high voltage packs would be another great feature.

Dig servos are different. Power to weight ratio is an issue with these. A very powerful, lightweight servo would be great for dig.

[JavelinSST390]

The way I see it the servos aren't the problem for requiring the BEC... the Speed controls are.

The best ESC on the market (Castle Creations Mamba Max) has a SUPER WEAK BEC that pretty much makes the external BEC a requirement.

And most of the more popular ones aren;t that great, either. I personally am running an LRP A.I. Runner, which has a super-weak 6v 1amp BEC. I am only running this ESC because it is waterproof. (We run in water a LOT down here). If I use my winch servo at the same time as I try to steer, both pretty much stop working.

What you guys should look into making is a Crawler-oriented BRUSHED ESC (not all of us are brushless, nor can we afford to be, and I really don't want to pay for a brushless ESC when I don't own any brushless motors) with a hold-brake or drag-brake function, and a strong BEC capable of running TWO high-torque servos at the same time.

And make it affordable... under $80.

[TheHeadHunter]

Originally Posted by JavelinSST390

The way I see it the servos aren't the problem for requiring the BEC... the Speed controls are.

The best ESC on the market (Castle Creations Mamba Max) has a SUPER WEAK BEC that pretty much makes the external BEC a requirement.

And most of the more popular ones aren;t that great, either. I personally am running an LRP A.I. Runner, which has a super-weak 6v 1amp BEC. I am only running this ESC because it is waterproof. (We run in water a LOT down here). If I use my winch servo at the same time as I try to steer, both pretty much stop working.

What you guys should look into making is a Crawler-oriented BRUSHED ESC (not all of us are brushless, nor can we afford to be, and I really don't want to pay for a brushless ESC when I don't own any brushless motors) with a hold-brake or drag-brake function, and a strong BEC capable of running TWO high-torque servos at the same time.

And make it affordable... under $80.

x2 hallelujah!

[Tekin Prez]

3 options.

1 – Offer a more powerful bec in the speedo. We might be able to do this on the FxR if we put it in the taller FxRpro case and double up on the regulators. Typically this will not provide much improvement unless they are properly heatsinked to the pcb, but it seems we are amping out and not overheating so it may get it done. We will mod some units and get them to our team guys to try.

2 - Design a servo that does not draw so much power and still has enough torque, probably have to give up speed. Stall current will still be an issue in extreme situations and it sounds like speed is still good to have so this is really a comprimise and thats not cool.

3 – Design servos that connect directly to the battery and they can have all the power they want. I like this one since the power you are expecting from the bec is getting pretty large. However there are so many good servos already out there and it is hard to justify the development for specialty applications unless you already have a servo line to work from.

I think for now #1 is our short term path and easy to try.

Tekin Prez

[Rocksmith Eng.]

Originally Posted by TheHeadHunter

x2 hallelujah!

X3
Having a esc that does not require a external BEC would be my first goal. A nice 300+ servo with lower power requirements would be second. Unless of course you can design a servo that will run on 1amp, be 400+ of torque with a speed of .35 or so and have a full aluminum case for heat dissipation and strength.

[JavelinSST390]

Originally Posted by teamtekin

3 options.

1 – Offer a more powerful bec in the speedo. We might be able to do this on the FxR if we put it in the taller FxRpro case and double up on the regulators. Typically this will not provide much improvement unless they are properly heatsinked to the pcb, but it seems we are amping out and not overheating so it may get it done. We will mod some units and get them to our team guys to try.

2 - Design a servo that does not draw so much power and still has enough torque, probably have to give up speed. Stall current will still be an issue in extreme situations and it sounds like speed is still good to have so this is really a comprimise and thats not cool.

3 – Design servos that connect directly to the battery and they can have all the power they want. I like this one since the power you are expecting from the bec is getting pretty large. However there are so many good servos already out there and it is hard to justify the development for specialty applications unless you already have a servo line to work from.

I think for now #1 is our short term path and easy to try.

Tekin Prez

That's all swell, but you just made a $150 ESC into a $200 ESC.

That hardly falls into the range of "affordable." :-(

Now, I realize there's lots of variables in business and such, but, in my humble opinion, I think a better idea would be to take the existing Rebel 2, bump the bec amperage as high as you can make it, and add a drag brake function.

Just call it "The Rebel Crawler". I'd buy two.

[Tekin Prez]

The FxR is $150 retail, but sells for about $100. It is a very aggresively priced speedo. The Fx and the FxR are intended to be a great value to get people to try Tekin and see what we are all about. The Rebel is actually as expensive to build because of all the hand work and the buying pwer we have on all the parts we use in our new designs. The Rebel also has an unregulated BEC that would be a lot harder to beef. It would require a new pcb layout that would be hard to justify. The last of them are on the shelf and I doubt we will build them again. There is still plenty left and we put them a really cool new package, but it may be time to retire one of the greatest and longest running speedos of all time!

We can do the BEC pump up for not much cost if it works well enough to justify it. A longer header pin, a couple of regulators and a little hand work. We might even sell it for only slighty more just because we like you guys! If we have to do a new pcb and a new case mold, then you are going to have to pay

Tekin Prez

[dr.chop]

I have tried a few servos, not so much in my crawler since I am still new to this, but in my Kyosho Buggy, and I am happy with the Ace Servos. They are way underrated, and have a nice silver heavy finned heatsink vs. the traditional black, dark, low finned, or smooth heatsink. They by far are cheaper new than most other comparable servos used prices. Digital, fast, and torquey is good. Plus, they aren't noisy like the 5955 I have. The nice part for me, is my radio lets me dial up or down the speed of the servo, so I am not too concerned if it is fast out of the box.

I am no engineer by any means, but I have noticed my 5955 has gotten very hot, as well has my son's 94357. I can only imagine that this summer, here in So Cal, we may run in to problems with the heat, and that should be addressed in design. As far as the BEC is concerned, I don't mind running the external CC on my Mamaba Max since it was a $20 add. Making a $200 speedcontrol doesn't really seem sensible in that aspect, and if a bec fails, you can easily replace it without being down and out. Servo construction is hot right now. Build a solid servo and it will carry over to other markets. Tekin has been a solid company in the past, and I think they will pull this one off well.