This article may be of interest to you if you have experienced either of the following problems with your mount:
- When you are moving the mount small distances using the slow-motion controls (either manual knobs or the hand controller for electric motors), if you change the direction of motion there is a noticeable and annoying delay before the scope responds and begins to move in the other direction.
- If you have a “go-to” equipped mount, you cannot seem to do the initial 2- or 3- star alignment accurately enough to result in accurate go-tos.
The first of these problems, and possibly the second, is a direct result of a mechanical property of the gears in your mount, called backlash.
This article discusses backlash and what causes it. You may be able to reduce somewhat the backlash in your mount, but you will not be able to make it go away entirely. However, by understanding what is going on, you can take corrective measures to minimize its effect on you.
What Do You Need Gears For?
If backlash is an undesirable property of gears, it’s natural to ask why we need gears anyway. Wouldn’t it be better to just eliminate them and the problems they cause?
Your mount is dependent on gears because it needs to make a shaft rotate around a centre point very slowly and very accurately.
Consider, for example, the rotation that the Right Ascension axis of an Equatorial Mount needs to track an object as the Earth rotates. The RA axis needs to rotate at the same rate as the Earth. That’s a rotation rate of “one rotation per day” or, in more familiar engine terms, 0.0007 RPM. That’s pretty slow.
Unfortunately, it’s quite difficult to build an electric motor, or a human wrist, that can turn that slowly and still turn at an accurate and consistent pace.
Instead, the normal engineering solution is to use a motor that runs quickly enough that a small error is insignificant (several hundred RPM) and to use a series of gears to mechanically reduce the speed to the needed 1-revolution-per-day.
For example, you might use a 300 RPM motor and a set of gears to reduce the rotation rate by a factor of 432,000 times to get a shaft rotating at a very accurate one revolution per day. A worm gear combined with one or two round gears can easily produce such a reduction in rotation speed.
Why Do Gears Have Backlash?
The need for gears seems pretty straight forward and, in theory, you would expect them to work perfectly. The idea is simple: two wheels of different sizes, with identical-sized teeth cut into them. The motor drives the smaller gear and its teeth, engaged with the teeth of the larger gear, drive the larger gear. The speed of the second shaft is slower than the speed of the driven shaft by a factor of the ratio of the number of teeth on the two wheels. (A 10-tooth gear driving a 20-tooth gear cuts rotation speed in half.)
The problem is the precision required for this to work. If any of the gear teeth are cut imperfectly – too large or with the wrong angle – the two gears will at least wear badly or at worst will actually jam when the over-large section tries to pass by its neighbour.
It’s certainly possible to machine gears so they mesh absolutely precisely, but it’s expensive. (The drive motor for my little M4 equatorial mount cost only $50, and it wouldn’t sell well if it had to be $400 to contain high-precision gears.)
Fortunately, there is a cheaper solution. If the teeth on the gears are made slightly smaller than they need to be, or are spaced slightly farther apart than they need to be, then the air space around the teeth provides relief so that minor imperfections don’t jam up the works. And this deliberate under-sizing doesn’t affect accuracy at all, as long as the motor is always being driven in a consistent direction.
Since most small electric motors are, in fact, used in applications where they are expected to run in one direction only, this technique is very common in the manufacture of small motors.
The animation above shows this technique. The space between the teeth is exaggerated, but you can see that the drive is smooth in its constant direction and that there is room for minor imperfections in the gears. You can also see the speed reduction – the blue gear is being driven by a motor, and the red gear’s shaft is turning at 8 / 20 = 0.40 times the speed of the motor-driven shaft.
The problem arises in an application where you need to be able to change direction. In this example, we periodically reverse the direction of the motor, and you can see that the tooth on the blue gear has to cross that little air gap before it presses on the tooth in the red gear that will be engaged in that direction. That motion of the small gear across the gap, motion that is not producing motion in the large gear, is backlash.
If you watch carefully, you can see that every time the motor reverses direction, there is a backlash pause before the large gear begins moving.
Dealing With Backlash
Backlash is annoying when you are trying to precisely guide your telescope to a specific spot, and it introduces error in the alignment of a go-to system. So what can be done about it?
There are two options: you can try to reduce or eliminate the backlash by improving the mechanical precision of the system, and you can accept a certain amount of backlash and use techniques to reduce its effects.
Some mounts allow you to adjust the alignment and spacing of their internal gears, giving you some ability to reduce backlash. Generally this is possible on higher-end mounts where the gearing is built in to the mount, but not on lower-end mounts where the gearing is an external after-market accessory. For example, gearing is adjustable for backlash on the Celestron CG5 and CGE mounts, the Losmandy mounts, the Atlas, etc.; but it is not adjustable (as far as I can figure out) on the accessory motor on my little Stellarvue M4 or the Orion Skyview Pro.
Adjusting gearing to reduce backlash involves some disassembly of the mount and is not a simple exercise (and I don’t plan to detail how to do it here). If you’re the kind of person who tinkers with their car or lawn mower engine just for the fun of it, you’ll probably enjoy tinkering with your mount’s gearing; otherwise it’s probably not worth it, especially since you can use simpler techniques to live with backlash (discussed below).
If you decide to tackle fine-tuning your mount’s gearing, you’ll find there are two sources of backlash, and thus two approaches to reducing it. They are related.
- Your mount may have imprecisely manufactured gears (rough surfaces, jagged edges, etc.). You can replace such gears with high-quality commercial ones, if you can find a supply of the appropriate parts. Or you can improve the precision of the existing gears by removing them and carefully sanding and polishing the rough surfaces.
- The gears may be mounted with too much free play between them. On most higher-end mounts you can finely adjust the position of the main worm gear and the toothed gear it drives, usually by adjusting some set screws, and possibly by adding shims. Be careful, though. As you mesh the gears closer together, the chances that they will bind increase. If you adjust the gears too tightly together, the mount will be hard to move, and that may put too much strain on your drive motors, causing them to run imprecisely or overloading them. That’s why you would normally work on improving the precision of the gears first, then on the precision of their alignment.
If you take apart one of the mid-range mounts (e.g. my former Celestron/Synta CG5) you’ll discover the gears are running in a coating of grease that is quite viscous (thick). The thickness of that grease is deliberate – it provides support and mechanical stability for gears that are fitted rather loosely together. Unfortunately that trick becomes a problem in cold climates like Canada – the grease thickens almost to the point of being gluey. Many people replace the grease with one rated for lower temperatures; but that may also require that you polish the gears and adjust the tolerances to make up for the extra mechanical play allowed by the thinner grease.
Google will quickly locate ‘net instructions for adjusting the gearing on most mounts, and I’m not going to repeat it here.
Compensate For It
Whether you adjust your mount to improve the mechanics or not, you will always end up with a certain amount of backlash, and there are things you can do to make it less of a problem.
The easiest thing to do is “get over it”. If you’re just moving your telescope around for short viewing sessions, backlash is not going to be a major problem and, now that you know it does not indicate something is broken in your mount, you can probably learn to ignore it.
Also under the “easy” heading, if you have a manual mount with flex shafts for slow-motion control, you should periodically check the thumbscrews where the flex shafts attach to the mount. These tend to work loose, and introduce a very large backlash between the shaft and the mount that has nothing to do with the internal gearing.
Extra Steps for Motorized Mounts
For equatorial mounts with built-in motors on both the Dec and RA axes (both go-to mounts and non-go-to but motorized mounts) there are some additional habits you can form, when using the electronic control pad, to reduce the impact of backlash.
Backlash in the Declination drive hurts nothing, it’s just annoying. The Dec drive is used only when you are locating and centring an object, not for tracking it. Just get used to the fact that, when you change direction, there will be a small delay before the scope begins to move.
On some electronically controlled mounts, the control unit has an optional feature that will cause the mount to “take up the slack” when you change directions, usually only on the Dec axis. Since backlash only happens when you change directions, the mount can be programmed to do a very short burst of fast motor motion before running at the regular speed, whenever you change DEC direction. You adjust this setting, using the mount’s control panel, so the length of the burst of fast motion is just sufficient to cross the gap in the gears that cause the backlash. This option is not usually applied to the RA motor, since that motor runs constantly on a tracking drive.
Right Ascension Backlash
Backlash in the RA axis, on the other hand, can actually interfere with your use of the mount, because the RA motor is used to keep objects stationary, compensating for the rotation of the Earth, after you have centred them in the field of view.
Let’s work through an example. Suppose you are observing Saturn at moderately high power (about 200x in this simulation).
Of course, in a non-driven mount, Saturn will drift out of your field of view, toward the West, as the Earth rotates. That’s why you selected a mount with a built-in motor drive on the Right Ascension axis. On a properly polar-aligned mount, the motor will exactly counter the rotation of the Earth and keep Saturn stationary in the field.
Now let’s assume you are using an electronically controlled mount with the motor drive running, and that you are using the East/West buttons on the electronic hand control to slew Saturn to the centre of the field.
If you happen to be approaching Saturn from the East, you may see an effect something like that shown in this simulation. You can see Saturn scrolling into the field of view as you activate the RA drive. Once it is centred, you release the RA button and allow the “tracking rate” on the motor to continue to run. But look what happens – Saturn drifts partway out of the field of view before it becomes stationary.
What’s happening? Backlash.
You approached Saturn by running the RA drive in one direction, then the regular tracking rate took over, but that requires the motor to run in the opposite direction. While the teeth were crossing the “backlash space”, the telescope tube was not being driven, and this gave Saturn some time to drift a bit. Once the offending teeth “crossed the gap” and were pressing on the other side of the other gear, smooth motion resumed and Saturn appeared to become stationary. It’s very annoying though – you took pain to centre Saturn in the field, and now it’s tracking but off-centre.
Is there anything you can do so that your target stays in the centre of the field when you stop slewing?
Of course you could “aim to the right” – try to guess where to let go of the button so the target drifts a bit and ends up centred. But that’s imprecise and feels wrong – “why should I have to try to outsmart my expensive new mount?”
Here is the technique: always do the final approach to the target with the correct gear teeth already engaged. For an equatorial mount, that means using the other button to do the final approach to the target.
So you’ll use the West button to approach the target, as before, but you’ll deliberately overshoot – go a little past the target. Then use the East button to back up until the target is centred.
Now when you let go of the button, the appropriate teeth are already in contact in the gears, so there is no period of backlash and no drift – the target remains perfectly centred and stationary.
Some go-to mounts do this automatically when they approach a target. I remember my Celestron CG5 AS-GT mount, for example, used to do it. I would occasionally notice it “overshoot” a target and then back up. On paying closer attention, I realized this was only happening with the Right Ascension direction, and only when approaching the target from the wrong side. The mount was ensuring that the final approach to the target was always with the correct gear teeth engaged, so there would be no backlash drift when slewing stopped.
Backlash and Go-To Alignment
A final place where you can take action to minimize the effect of backlash is if you have a go-to equipped equatorial mount. As part of setting up such a mount, you will have to do a 2- or 3- star alignment, where the mount slews to the expected location of some bright star and then asks you to perfect the centring manually to calibrate the go-to model.
When you are using the hand controls to precisely align the chosen target stars, think about what happens if you change direction, introducing backlash. The go-to computer will note that the direction button was pushed, but the telescope will not have moved the appropriate amount because of the backlash delay. This means the distance information the go-to computer uses to calibrate will be off by a small amount. The result will be an imperfect sky model, and go-to operations that are “close but not quite right”.
What to do?
This is simple to resolve. Pick a direction: up or down, and another: left or right, and always do the final approach to your alignment stars using these buttons on the hand controller. (The Celestron manual recommends always using “up” and “right”). By always approaching the alignment stars with the same buttons you will not be sending alignment information to the computer in which backlash has introduced errors. The result will be go-to operations that are far more accurate.
So, let’s summarize that. When centring alignment stars during go-to calibration, always make the final approach to the alignment star using the Up and Right buttons on the hand control. Overshoot and back up if necessary, but make sure the last buttons pressed on the hand controller are Up or Right, not Left or Down.
Gears are a natural part of telescope mount mechanics, and backlash is a natural property of gears. Although you can reduce it by careful assembly and adjustment, it will never be zero. However, understanding what is going on and forming some simple habits will minimize its negative impacts on your observing experience.