Cheap servos worth the risk?

I'm about to buy some replacement servos for my ongoing repair project and I'm intending to upgrade to metal gears whilst I'm at it.

Looking around has revealed some cheap servos available, as little as £4 each, with metal gears. However, I'm sceptical about the quality at that price and if I'd really want to trust an aircraft under my control to have substandard parts that could cause an incident.

Are my fears warranted or can cheap servos be just as reliable as their more expensive counterparts? My mind is currently swayed towards sticking with a trusted supplier and manufacturer and spending that little bit more, hoping it buys me reliability when I need it most.

Have you bought cheap servos? Did they perform well?

depends on the model. Clearly an 80 inch warbird needs better quality gear than a foamy.

I dont use anything other than futaba, savox and hitec but in truth recent issues have made me question my use of hitec in some respects.

We have a test model here at laser for flying engine prototypes and when setting it up i originally got some of the ripmax newpower servos for it to save a few quid. I ended up sending all 5 back as they were very poor in terms of accuracy and it was very easy to send them into some sort of fit from which they would never recover and just got really hot.

I dont think you need to go mad on servos, but dont penny pinch too much either.

As a general rule of thumb, cheaper metal gear servos tend to suffer from more backlash than nylon or karbonite gears. Personally I'd sooner spend the extra money on digital servos rather than metal gears. Like Jon, if I want peace of mind I buy Savöx for my fast or more valuable models, otherwise I use Hitec or SD-100/150 servos. Generally you only get what you pay for so a cheap servo is going to be a bit of a lottery with reliability and who knows if the quoted torque & speed specifications are correct or not. Probably not.

The old adage of "buy cheap, buy twice" springs to mind.

From personal experience, cheap servos work okay in applications which are not too stressful. . The performance figures quoted on the box can be somewhat optimistic, and the centring can be a bit vague at times. . Avoid putting them on elevators in a power model, especially if the engine is powerful as the draught (thrust) from the prop can create difficulties for "low rent" servos.

As Jon says, some Hitec seem to have a short lifespan these days too.

Cheapy servos are okay in foamies and trainers but, for more serious models, it is worth going for better quality equipment.

B.C.

How about trying to fly your model with a failed servo and see how well you cope (simulators provide a random option) and see how you get on?

Just get 4 mistakes high (one extra for luck!) and hold a stick hard over, then fly about and land when ready. If you get into trouble just assume normal control..

Practice will make you ready for nearly any eventuality whether its a cheap or expensive servo failure!

PS £4 for a servo so the retailer buys it for £3 and the manufacturer sells for £2 and it costs them £1 to make and your model is worth?

I have never experienced a servo failure in flight, and I've used a wide variety of servos, both expensive and cheap.

As Jon says, it really depends upon the suitability of servos for the model, rather than branding each offering 'good' or 'bad' in its own right.

Just one flip side of the coin, though. I tend to buy spares (to make the most of postage costs) when I buy servos, and I have swapped out 'suspect' servos that have become glitch or noisy in ground testing long before they have a chance to fail in the air.

It's possible that I'd be less likely to either buy lots of spares, or swap out suspect servos, if those servos were very expensive ones. So there's a balance. Yes, you often (but not always) get what you pay for. But buying expensive does not automatically guarantee against failure.

Edited By The Wright Stuff on 28/06/2017 09:18:25

I bought a servo tester from HK at the same time bought 8 metal geared corona digital servos. Put them on the servo tester and left them running (3 at a time)

3 were warmer than the rest so got them replaced by HK, new ones are ok. Generally any faults show up on the servo tester after 15 mins, I allowed about that.

All servos were under a tenner each and I now buy corona with confidence (every bubble's passed its fizzical - if you can remember that ad you're as old as me ) Corny joke aside I now have these digital metal geared servos in my petrol powered extrawot, YT dragon lady, seagull spacewalker, maxford Cub and cannot fault them. In the smaller sizes, 18 gm and 22 gm, the gear train is much more reliable than than sd 200s. The only long term problem with them is getting new servo arms, when it is almost as cheap to buy a new servo.

HTH.

PS use thread lock on the servo arm retention screws.

But never on plastic/nylon or carbonate gears.

Bear in mind that there are servos that are relatively low cost and really good value for money and then there are cheap servos in every sense of the word and the art is differentiating between them.

Edited By Andrew Ray on 28/06/2017 09:50:58

There are some real rubbish servos sold on the net ,beware! Servos being  sold as Futaba are often poor copies and not fit for purpose. If price looks too good to be true it probably is.  Buying good secondhand is far better than buying these cheap copies. You can get reasonable priced servos but buy from reputable supplier like Nexus or RC Life etc or you LMS . Tower pro used to be OK ish but quality varied and even they are being copied and poor copies at that. 

" I'm sceptical about the quality at that price and if I'd really want to trust an aircraft under my control to have substandard parts that could cause an incident"

£4 servos for a 1lb £40 foamie airframe, fine, I guess. Even then, from my sample size of 4x9g servos, 2 failed within a couple of flights (rudder stuck hard over, one aileron stuck half up), and I've subsequently replaced all of them with HS55s, losing two was more than enough for me.

On a £300 IC model, IMO I wouldn't touch the cheapies - as said, buy cheap, buy twice, but in doing so you're likely to trash an airframe in the process.

Some things are worth sticking to a reputable brand, from a normal retailer.

"As Jon says, some Hitec seem to have a short lifespan these days too."

chaps, which hitecs have you had issues with?

225s have a well documented weak gear, beyond that I'd always thought of hitec as a fairly reliable servo?

By far my worst experience has been with SD200s - a supposedly "decent" quality servo. I've had two or three fail in service - and one that drove to the far end when I connected it, straight out of the box! The other common source of problems I've encountered has been pre-installed 9g ones in foamy ARTFs.

When building a new larger and/or valuable model, is it better to use new servos or ones that have had a reasonable "burn in" time in a hack model? Maybe an extended test period on a test rig would be a compromise but without realistic loads and vibration, maybe not as good as in a model?

I swapped out a load of HS-300 standard servos because I wasn't 100% happy with them, but they were over 10 years old at the time. There was a strange latency: a delay of about 1/4 second between moving the sticks and moving the servo. It disappeared when I replaced them with HS-322.

In hindsight, it could have been resistance in the plugs / wiring due to corrosion, I suppose...

There used to be kit for life critical applications, 10 hours of use then sale as fully tested. And expensive.

Servos are similar. But bets off after an arrival. I have always liked plastic servos where appropriate. Accepted high vibration models and big models like metal gears better.

But if an arrival strips a plastic gear, that is a clue you have overstressed it. With a stronger metal gear, are you sure something else hasn't shifted, bent or cracked. And the control arm won't fall off in the absence of thread lock on a plastic job.

So a question to the original poster, why is an upgrade needed. What was wrong the the old spec servo.

On what I find OK, Turnigy, Corona, Futaba. But I have had a few issues with standard Futaba S3152 digitals.

Nigel... To be fair, I fly a lot and my style of flying goes somewhat beyond genteel circuits with an occasional loop and roll thrown in to break the boredom. Over the last 50+ years, I have had 14 servos (of various manufacturers) fail in flight. These are usually elevator servos but, fortunately, there are usually two servos employed so it doesn't end in a disaster. . I have lost count of the number of servos which have been replaced because they went "moody" but did not get the chance to fail in flight. . Again, various manufacturers, various qualities but, ultimately, all bin jobs.

At the risk of taking this thread off topic, I too had assumed that Hitec servos were fairly reliable but experience has shown the ones at the bottom end of the range to be a bit "fragile". . . I have no desire to get into a disagreement with anyone who wants to defend them, but these are my experiences with Hitecs from the basement end of the range:

HS-55 Feather servo: After a short time the resolution goes sloppy. After that, they go slow and burn out.

HS-81 and HS-82mg: They don't stand the test of time and will run slow (or run slower in one direction than the other) as a prelude to burning out. . If used on a throttle, it is amazing how long the fuel lasts when one of these fails in flight..

HS-125 (slim): Not enough power to deploy a flying surface -- it blows back. The specs on the box bear no resemblance to performance in the real world.

HS-225: The gears go sloppy after a short time.

Basically, these tinier Hitecs don't seem to like running on 6 volts . . but who runs anything less these days?

However, Hitec servos in the Premium (top end) range seem to be great. They deliver excellent performance and seem to be reluctant to wear out even in 3D models powered with engines north of 60cc . . . and they can comfortably handle the volts from a 2s Lipo.

Oh, and their retract servos are great too. Smooth, powerful, they go at the right speed and they last for years.

B.C.

I have used hitec 311's for some time but now have two in my hurricane where their centre point wanders about.

The issue is not that the dont centre, its that the point at which they centre can change day to day and i have to sub trim it out. Usually after a very exciting first flight of the day.

Its really annoying

Wow! Lots of great responses again! Thank you all again for your support, it really does help me a lot!

Donald, in answer to your question, my thinking is that nylon is a bit weaker / cheaper and thought the only reason for that was to be cheap and maybe lose a gram in weight which my trainer won't be too bothered about. I'm also considering that sometime in the future I'm going to want to recycle this model (meaning its probably going to see the ground coming up a bit too quick, just once too often) and then I'll have good solid servos that I can re-purpose into more challenging models later. Thinking its better to spend a few extra quid now to have that ability later.

More than happy to be corrected, I'm a total novice at all this and looking to learn from other peoples' experiences.

thanks again

Tony

thanks - seems the cheaper hitec stuff is where the problems are, by all accounts

I don't think we're off topic talking about hitec servos, as they do make some very cheap servos.

I have used quite a few Corona DS939MG 12g servos in small models with good effect, finding them reasonably accurate and fast so good value. I generally use Hitec servos although I have had a couple fail out of the box but not inflight. One for a TREX 500 that smoked when I plugged it in and another for a DLG that was completely random in its movements. Both were replaced by return. Getting what you pay for is probably true but I connect any servo I buy to a tester, cycling for some minutes. The theory being that if it is a 'Friday afternoon job' it is probably going to fail sooner rather than later.