Gasser throttle servo?

by | |

Q. Throttle servo on my gasoline powered model quit. Flew 10 minutes at part throttle until the engine died (so I could dead stick to landing because the throttle cut wasn't working either). Glad I didn't crash! Anyway, since I don't want a repeat, which throttle servo would you recommend for a gasser?

A. The answer is . . . it depends. In general standard size servos work better than mini or micro servos because they are more massive and better absorb vibration. Engine vibration is a big deal. It's what often kills throttle servos. And with a single-cylinder gasser, especially, because they're akin to a paint shaker and are rough on servos, bigger is better! Note; my guess regarding what killed your throttle servo, and without seeing the servo or even knowing which brand, is that the potentiometer failed. Usual suspect.

So what are factors at work when selecting a throttle servo for gasser powered project? Cost is a big one, so is performance, e.g. aerobatic models favor faster servos. But faster servos cost more money. So like I said, the answer to which throttle servo is best is complicated. Let's delve into the subject by looking at examples.

For a warbird like my 85cc powered Mustang? I'm perfectly satisfied with a DS90DLHV for throttle (and choke). And I like using a PDRS105 HD arm because it's perfect for a 4-40 ball.

What about aerobatic models. For example, a 74" Extra powered by a DL35, especially if you have aspirations of 3D, then the DS140DLHV is a good choice because it's faster. 

However, a better servo for any 3D model Yes, including 170cc powered 40% class), would be a DS305CLHV. Not because it's faster than the DS140 but  because it has steel gears. Costs more but what price performance?

Gear material is a very important factor. First, it's because steel is the strongest gear material, but second, because it's steel is also the most durable in a high vibration environment.

Vibration considerations are an especially big deal with single-cylinder engines. So what's our best throttle servo? Our best is the DS555BLHV.

What about for helicopter's, especially ones equipped with Zenoah single-cylinder thumper? Again, the DS305CLHV or DS555BLHV are favored for their steel gears.

Boats? Boats are HARD on throttle servos! Partly it's a matter of proximity. But shock plays a huge role because at 70mph water is hard. Honestly, all the factors mentioned above plus mounting location and pushrod material are going to be worth paying close attention to details (more below). Note; lot of boat guys like the DS90DLHV because they're cheap and  pretty good but because of the steel gears, I'd be looking hard at a DS305CLHV, again.

Harder than pretty much anything on throttle servos are the 35cc powered 1/5th scale gas trucks. Honestly, think of it as a helicopter on wheels but with harder G-forces playing a role than with boats because rocks, roots, and fence posts are murderous. Again, pushrod materials are important but so is using a polymer servo arm like the PDRS105 for it's vibration absorbing capability.


Servo leads have three wires, two (the +, which is orange and -, which is brown) are for DC-power, and the third wire (yellow) is for the pulse. The width of the pulse is what determines the angular position of the servo. Changing the width of the pulse changes the position.

- Three wires, plus and minus for DC-power plus the pulse wire

The range of pulses for a servo connected to a receiver is typically between 1000-2000µs. The center of travel will be in the middle of this range, say 1500µs. Thus, low throttle may be 1000µs, mid-stick is 1500µs, and high throttle is 2000µs. Changing the pulse width by moving the throttle stick is called modulating the pulse, or changing the pulse width. Does it follow that 1/4-throttle will be about 1250µs while 3/4-throttle will be 1750µs? Yes, that's it exactly!

Anyway, the technical term for this is PWM (for pulse width modulation), and the same holds true for every channel, e.g. elevator, aileron, etc. To recap; when you change the width of the pulse by moving the control stick, you change the position of the servo.

The important thing to realize is this; servos designed to be connected directly to the receiver have a range of between 1000-2000µs. And whether it's full left to full right, or low throttle to high throttle, roughly 1000-2000µs is the range of pulses they're designed to accept. You find this information on the specifications tab for every servo - like the one below.

- There's lots of good information to be gleamed by reading the specs

Importantly, servos are defined by type and this is based on their neutral position, or 1520µs for our servos (e.g. their midpoint). Also, you may see some brands of servo where the the midpoint as 1500µs instead of 1520µs but this is essentially the same thing (and plenty close enough for our purposes). E.g. this detail doesn't matter.

Tail Rotor Servos

Meanwhile, there are a class of servos, which we call tail rotor servos. Their pulse range is half as wide (the why is beyond the scope of this article so please just accept that they're different). Anyway, for these servos, instead of neutral being 1520µs, it's half that, or 760µs.

Thing is, they're fast and we've had folks buy them by mistake and have to return them. This is because instead of their range being 1000-2000µs it's about half, or 500-1000µs. Anyway, since their range isn't within the range of the signal sent by the receiver, tail rotor servos require a gyroscope (connected between servo and receiver) to re-interpret the 760µs neutral to 1520µs (so they'll work with a receiver).

The point being, without a gyro, a tail rotor servo won't work with a receiver. And note; while plugging a tail rotor servo directly into a receiver won't work, it also won't hurt it because the receiver doesn't even know it's there! Anyway, we know all this from the specifications for the pulse - see below.

So how do you tell the difference? You look at the specs! Above is a detail from the specs for the DS120BLHV tail rotor servo. There are three lines of special interest right now.

The first is for max travel (depending on the servo, this ranges between 110-125° or 400-1100µs in the case of our tail rotor servo). Max means max, e.g. the very very maximum it can sweep - i.e. if you max out the travel volume of the particular channel on the transmitter by setting it to 125% or 140% each way. Note; this varies by brand of transmitter with Futaba maxing travel at 125% and others like Spektrum maxing at 140% - but - the end result is the max the servo can accept is exactly the same for all brands of transmitters because it's determined by the servo. Anyway, the second line is the normal travel, which is 90°, and the third line is for 0° (neutral, or center).

The take away from all this is 760us is outside the typical range for a receiver and thus, tail rotor servos won't work when plugged directly into a receiver for throttle use. This means you can forget about using a tail rotor servo for the throttle or anything that's connected directly to a receiver because they don't speak the same language and must have a gyro.

Throttle servos

Now let's get to the meat of the question; selecting a throttle servo for gassers - and also - whether you can use a mini instead of a standard-size servo.

First, it's important to understand the servo doesn't know (or care) what use you're making of it. E.g. whether it's function is controlling the throttle, elevator, or rudder - it simply doesn't matter. All it 'knows' is the pulse width. Does this mean you can use a mini instead of a standard size servo? Yes, absolutely. The real question is . . . should you? The answer is - it depends.

Here's the rub with a servo, any servo - ours, or another manufacturer's, for some reason there is a fondness amongst giant scale modelers for placing the servo directly on the engine box - and - connecting it to the carburetor with a short straight steel linkage rod. This is bad for two reasons.

- This is the worst possible connection . . . a direct steel pushrod!

The above photo shows the worst connection you can make between servo and carburetor, and it's a bad idea on multiple fronts. Partly because mounting the servo on the engine box places it directly on the highest vibration surface of the whole aircraft. The why for this involves model designers and their foibles, which is above my pay grade. However, 'if' I had my druthers I'd never mount a servo on an engine box. Not ever!

Anyway, below is a cross section of a typical ProModeler servo. And please note; this isn't an unusual configuration for the potentiometer. In fact, it's probably what you'll find in every other servo on the market! Do you see how it sits directly below the output gear, which is made of metal? If you connect something vibrating like mad to the output gear, the vibrations feed directly into the pot. There's nothing we can do about it as a practical matter. So we have to deal with the vibration some other way. And note; this is not a defect in design, it's just how you make the servo compact enough to be of practical use. Put another way, it is what it is.

- The potentiometer directly engages the output gear - no isolation at all.

Anyway, the pot sits directly below gear and is connected mechanically to the output shaft on which we install the servo arm. So whenever a modeler uses a steel pushrod for the throttle pushrod - the engine vibrations are being transmitted directly into the potentiometer. This is bad and means the pot's lifespan is being degraded much faster than would otherwise be the case. How much? From thousands of flights to dozens. Sometimes just one!

"But ProModeler servos meet MIL-STDS for vibration" you say! Yes, this is true. But the test is for the electronic components mounted to the PCB. What modelers are doing is totally different because they're sometimes using a short steel rod attached to the steel arm of the carburetor (of a vibrating engine and often with an aluminum servo arm on the output shaft of the servo). It's almost as if they wanted to destroy the function of the servo - not kidding!

Basically, the steel rod feeds destructive vibrations directly into the potentiometer itself. Worse, steel does absolutely 'nothing' to attenuate the vibration. Frankly, it's no wonder the throttle servos may quit working in short order with gassers. And it's entirely to do with the potentiometer being eaten alive by the engine vibration!

Fortunately, if your servo dies, it's probably just the potentiometer and we offer replacement potentiometers. Moreover, they're mostly easy to replace (three solder joints) so it's a DIY for a lot of modelers. The upshot is; unlike other brands, because we'll sell you just the pot you can fix it yourself instead of being forced to send it in for repair. The other guys? It pays to ask about this ahead of time, capice?

- Killing a potentiometer isn't the end of the world - it's an easy fix.

Is there a solution for dealing with engine vibration? Well, yes, there is. This involves a) not using a steel pushrod and b) not mounting the servo on the engine box, and c) using a plastic servo arm.

Since modelers are going to persist in mounting the servo on the engine box, the next best thing is to use a flexible pushrod to connect carburetor and throttle servo, e.g. eliminate the steel pushrod. However, we also counsel mounting the throttle servo for our gassers at the CG of the model and using a Gold-N-Rod to make the connection as well. Yes, this means a pushrod two feet long. So what? It doesn't matter in the least! We also prefer using a polymer servo arm instead of an aluminum one for the same reason; plastic attenuates the vibration better. Note, Sullivan sell these with red, blue, and black sheaths. In my experience, the blue sheath stuff is usually good enough (the black is designed for 4-40 hardware while red and blue are going to use 2-56 hardware).

Note; there are some added benefits of doing it this way. First, with the servo at the CG, the model's structure attenuates lot of the vibration from the engine. Second, not having a steel pushrod eliminates the worst of the vibration from being transmitted into the pot (plus the possibility of ignition noise making its way into the servo and driving it crazy). Third, the plastic servo arms attenuates some of the vibration as well. Fourth, the model tumbles about the CG more easily because the mass of the servo is right on the CG (remember the dumbbell effect). However, and in all honesty, only the best pilots can possibly tell the difference involving moving a few ounces to the CG - but our view is every bit helps.

Thus, our recommendation is - at a minimum - using a short section of the inner part of a Sullivan Gold-N-Rod instead of a steel pushrod. Why? Principally because the plastic pushrod helps attenuate vibration enough to let the servo live. And while a plastic servo arm also helps, the biggie in all this is the plastic throttle pushrod.

- Best practice involves a flexible piece of Sullivan Gold-N-Rod inner.


- SolidWorks render of DS345BLHV showing component layout

Final thoughts

Some internet expert always brings up expansion/contraction of a Gold-N-Rod as a reason to not use it. As if steel doesn't expand and contract also, duh! Anyway, this is totally a non-issue, believe me. Remember, when dealing with a paint shaker of a gasser engine, the goal is to attenuate vibration and your servo will live a long life regardless of whether it's a standard size or mini.

This entry was posted in no categories.