Fixing declination runaway on a Meade LX200 telescope

[collinsmark]

TL;DR Summary

This thread documents the repairs I made to my 10" LX2000-ACF telescope to fix a declination runaway problem. It involves replacing an optical encoder attached to the declination/altitude motor.

Part I: Introduction to the problem.

I made this thread to document and discuss the repairs I made to my 10" LX2000-ACF telescope to fix a declination runaway problem. It involves replacing an optical encoder attached to the declination/altitude motor. I expect this documentation will be broken up into around 5 parts or so.

What is "declination/altitude runaway" or "right ascension (RA)/azimuth runaway"?

It means the telescope moves along a given axis at high speed, seemingly on its own volition, when it shouldn't, even when the telescope is properly aligned. Worst case, this can cause all sorts of problems, including equipment damage.

What "declination/altitude runaway" or "right ascension (RA)/azimuth runaway" is not:

If your scope is a GoTo scope, it needs to be properly aligned. Lack of star alignment can cause the telescope to point in the wrong direction and slew to the wrong part of the sky. That's not a "runaway" problem, that's just poor alignment. The rest of this post assumes that at least some effort has gone into aligning the telescope mount.

It's also not user error. If you tell your scope to point to the middle of the Earth using the hand controller, or if the computer program you're using to control your scope tells it to do that, that's your fault. Don't do that.

What causes "runaway"?

It's almost always the optical encoder. Sure, it's technically possible that it's a circuit board failure, but it usually isn't. It's almost always the encoder.

What is an optical encoder?

An optical encoder is a device that provides feedback to indicate how far the shaft of a motor has turned. Generally there are two types:

• Relative encoders: These encoders simply enable a way of counting of how far a motor has turned clockwise or counter clockwise since some point in time, usually from the time the telescope is powered on. They don't provide a direct indication where the telescope is pointing. Their output output is relative to the point at which the counting started, and only can indicate how far the motor shaft has rotated, not necessarily the whole telescope. These types of encoders are relatively inexpensive. This is the type of encoder that will be discussed here.
• Absolute encoder: These types of encoders can give the system information about where exactly the telescope is pointing in absolute terms, even if the telescope is just powered on, and even if the user manually disengages the clutches and moves the scope around manually. These types of encoders are very, very expensive, and will not be discussed here.

Optical encoders involve an LED, a couple of photodetectors, and a wheel, as shown in Fig. 1:

Figure 1: General idea of operation of optical encoder (image courtesy of ForumAutomation.com)

Although not shown in Fig.1, there are may also be reflective elements such that the LED and two photodetectors can all be mounted on the same circuit board. That is the case with the encoders involved in this thread.

Note that because there are two signals, A and B, the control circuitry is able to determine not only how far the shaft has rotated, but also its direction (clockwise or counterclockwise).

What are the symptoms of encoder failure?

Early symptoms of encoder failure often involve pointing accuracy problems. Particularly, situations where the scope was pointing and tracking accurately, then quite suddenly, and inexplicably, points to some other part of the sky. This is not conclusive evidence of encoder failure though, since many other things can cause such symptoms.

Things to rule out first:

• Cable snags
• Loose clutch (where one of the clutches slips)
• Neighborhood badger battling telescope in a fight for dominance
• Temporary power outage
• The mechanism connecting the optical tube assembly (OTA) to the mount comes loose or slips
• A piece of your telescope or falls off (e.g., counterweight) suddenly changing the telescope's balance

One should rule out other causes before blaming the optical encoder.

Later symptoms of encoder failure involve runaway, as previously mentioned. When attempting to slew the telescope, if one of the axis motors slews "balls out," so-to-speak, at very high speeds, even if the slew rate is at one of its slower settings, that is a symptom of encoder failure.

How to conclude that encoder failure is the problem?

If you can reproduce the problem of slewing at a high speed, even if the slew rate is at one of the slower settings, bring up your mount's coordinate information: the coordinates the mount "thinks" it is pointing. If you slew the telescope along the axis in question (the one that goes "balls out") and the telescope's coordinates do not update, that's a (nearly) sure sign of encoder failure.

Fig. 2 shows this on my Meade LX200-GPS mount. To get to this screen on the hand controller, press the "Mode" button for several seconds.

Figure 2. Axis orientation displayed on LX200-GPS hand controller

In my case, the "Dec" did not update even though the telescope slewed like mad when adjusting along the declination adjustment controls. This is an almost certain indication of the optical encoder failure on the declination motor.

To be continued ...

 

[collinsmark]

Part II: Disassembly

The rest of this description is specific to my telescope, a Meade 10" LX200-ACF, circa 2007-2008-ish. Although my telescope is a 10-inch, I'm pretty sure the procedure would be exactly the same for the 10-inch, 12-inch, and 14-inch models.

This description only involves the declination/altitude motor encoder. If you were to replace the right ascension/azimuth motor encoder, you'd need to go in through the bottom of the telescope, and that is not shown here.

Tools and stuff:

You'll need a set of hex wrenches, a place to temporarily store several hex screws & a spring, and a some sort of clipper tool to clip off a cable tie. You also should have some cable ties for later, since we're going to be destroying one of them.

Figure 3. Tools and stuff.

Figure 4. Extra cable ties.

Fig. 4 shows a bag of extra cable ties. We won't need these for awhile, but we will need one before we're finished. They're very inexpensive. You can by them by the hundreds, costing just a little more than the cost of postage.

Edit: New Encoder.

You'll also need a new encoder of course. The new encoder I used was US Digital part number E4T-256-091-S-D-D-2. More on that in upcoming posts.

Preparation:

Figure 5. Give your telescope a loving hug now, and whisper soothing words to keep it relaxed. Also, make sure it's powered off.

Disassembly:

Figure 6: Declination clutch knob.

To remove the declination clutch, loosen the clutch knob, and then just keep going. Be careful though, because once the clutch knob is completely loose, things can slide off (see below). So make sure the forks are level.

Figure 7. Declination clutch knob removed.

Once the clutch knob is removed, set it and the little cylinder thing aside. Be careful with the little cylinder thing, it can slide right off the bolt. We'll come back to these shortly.

The next step is to remove the 6 hex screws holding on the cover.

Figure 8. Declination mechanism cover removed.

Again, make sure the forks are level. There's nothing securing the big part of the clutch and large gear at this stage, so be careful.

To keep things secure, screw the clutch knob and cylinder thing back in place -- without the cover -- just to keep things from falling apart.

Figure 9. Clutch knob and cylinder thing temporarily reattached.

Take note of the cable tie and how it is affixed. We're going to cut that cable tie in order to gain access to a larger hex screw which is being blocked beneath the choke coil.

Figure 10. Declination motor assembly, still attached, prior to disassembly.

Figure 11. Cutting the cable tie.

Be careful not to accidentally cut any of the electrical wires.

Figure 12. Take note of the two larger hex screws holding on the declination motor assembly. We'll be coming back to those.

To be continued...

 

[collinsmark]

Part II (continued): Disassembly

Before we remove the declination motor assembly, we'll need to remove the declination adjustment knob first.

Figure 13. Declination adjustment knob removal.

Take note of the spring that's under compression, shown in Fig. 14. This is the spring that presses the worm gear up to the big gear that we mentioned earlier. It's an important spring. When you remove the declination motor assembly, it will slide right off. So make a note of where it is/was, and try not to step on it if it falls to the ground.

Figure 14. Declination motor assembly spring. Take note of this spring. Don't lose it when removing the declination motor assembly. You'll need to keep track of it when removing the assembly.

Finally we can remove those two larger hex screws that affix the declination motor assembly. (Again though, pay attention to that spring.)

Figure 15. Declination motor assembly removed.

By the way, I should have mentioned this before, but there's a lot of grease between the big gear and the worm gear. Avoid touching that grease. The grease is where it should be, and you don't want any on your hands because you'll probably want clean hands for what's to come in the next part.

Figure 16. Declination mechanism housing with the motor assembly removed.

The next part is to bring the motor assembly to the laboratory (or desk or countertop or whatever).

To be continued...

 

[collinsmark]

Part III: Old Encoder Removal

Now that the motor assembly is removed from the telescope, we'll need to prepare the old encoder itself for removal.

First off, Meade put a lot hot glue around the old encoder. I'm not quite sure why, but I figure they had their reasons. Much of the hot glue is concentrated around the motor's electrical contacts. So maybe it was to prevent corrosion or shorts involving the contacts.

We'll need to scrape off enough glue such that we can remove the lid/cap of the old encoder. Feel free to leave any hot glue that's not keeping the encoder lid/cap attached.

Figure 17. Old encoder with a bunch of hot glue on it.

I used an X-ACTO knife to scrape away much of the hot glue.

Figure 18. Old encoder with much of the hot glue cleared away.

In Fig. 18 you can see the old encoder's model number, US Digital E4P-256-091-L-S3039. As it happens, US Digital's E4P line of encoders are all discontinued. So we won't be replacing it with another encoder of the exact model. Instead, we will be replacing it with a similar E4T model. There will be a few differences between the old and new encoders.

Figure 19. Old encoder with cap removed.

Fig. 19 shows the encoder with the cap removed. Prominent in the image is the old encoder wheel. Notice that the side facing the camera is completely reflective.

Figure 20. Old encoder wheel and cap.

In Fig. 20, you might be able to make out a thin ring of stripes on the wheel on that side. The other side of the wheel is reflective though. Altogether this allows the LED and photodetectors to be placed on the same printed circuit board (PCB). Notice there's nothing special on the old encoder cap. It doesn't play a role in the encoder optics, it's just a cap. This will be a little different in the new encoder and new encoder cap.

Figure 21. Old encoder PCB.

Once the wheel is removed we can examine the old PCB. Notice a couple of screws here. The screws secure the PCB and the encoder base (we'll get to that in a second), into the motor housing itself. There will be differences in how all this fits together with the new encoder.

Figure 22. Old encoder base atop motor.

Removing the PCB we can see the encoder base as it sits atop the motor. The base contains a big hole in it, and that hole fits flush around a cylindrical extrusion (or circular "lip" if you'd rather -- I don't know what to call it) that's part of the motor housing. Things will fit together a little differently with the new encoder.

Figure 23. Motor housing with the old encoder (cap, wheel, PCB, base; i.e., the whole kit-and-kaboodle) removed.

Fig. 23 shows the motor housing with the encoder base removed too. Sorry about the bad focus. You can see that cylindrical extrusion/lip on the motor housing. While the old base had a hole in it that fit flush over the extrusion, the new encoder base will fit on top of it -- not through it. More on that in the next part.

To be continued...

 

[collinsmark]

Part IV: The New Encoder

Recall the old encoder was US Digital part number E4P-256-091-L-S3039. But the E4P line of encoders is discontinued. So instead, I used

New encoder: US Digital part number E4T-256-091-S-D-D-2

To explain the part number:
256 counts per revolution (CPR)
091 is the code for 2.3 mm bore size
"S" for Single-Ended
"D" for Default Cover
"D" for Default Base
2 is the code for Packaged Individually. (Unless you want to buy these in bulk.)

Don't do what I originally did and go to the large, online electronics distributor and start buying E4T encoders all willy-nilly. The other numbers in the part number are important.

Unfortunately at first, my online source for electronic parts did not stock US Digital part E4T-256-091-S-D-D-2.

Fortunately I gave US Digital a phone call (you can find their phone number on their website) and they were very nice. I was worried that they only sold in bulk. But they sold individual kits too! Oh, what a relief! Thanks goodness. They were extremely helpful and supportive. I bought 2 kits. One for this project and another in the event that I screwed up royally on this project. (I didn't screw up -- I'll use the other for the Right Ascension motor encoder at a later time, if that one ever starts acting up.)

Figure 24. New encoder in its packaging.

The new encoder kit comes with the following parts:
Cap
Wheel
PCB
Base
Metal, cylindrical spacer
Plastic, rectangular spacer
Two new screws

There's a video on US Digital's website on how to properly install these. I suggest reviewing that if you want to install one yourself (don't use my example, alone).

First things first. These optical encoders involve optics, so it's best to wear protective gloves to protect the optical components from my own grime and filth (See Fig. 25).

Figure 25. Powder-free vinyl gloves.

Figure 26. New encoder base vs. old encoder base.

Right off the bat, you'll see some differences between the old and new encoder parts. One difference is the hole size in the encoder base (Fig. 26). I panicked when I first noticed this. But before getting any ideas about getting out the power drill and trying to make a bigger hole, realize that the new solution doesn't really need a bigger hole. Notice the screw holes on the new base are recessed in a little (compared to the old base) and that should offset some of the necessary depth that the screw needs. Also, you're not screwing in the PCB with the new solution, so that gives you some more screw depth.

So in the end, I just let the new base sit atop the motor's cylindrical extrusion/lip, instead of around it. Everything seems to work OK with that solution.

Figure 27. Slide on temporary, cylindrical spacer to get the spacing correct when screwing in the new base.

When attaching the new base, slide on the cylindrical spacer over the shaft to ensure that the spacing is all correct when the base screws are screwed in.

It might help to have a third hand (Fig. 27) if one is available (one to hold the spacer, and the others to screw in the screws), but it can be done with just two hands.

Note: Do not screw in the PCB. Unlike the old encoder, the new PCB is not held in by the screws. The screws secure the base only.

By the way, I reused the old screws when attaching the new base. I know, the new kit comes with new screws, but the old ones were a little longer and they seemed to work fine for me.

Figure 28. New PCB installation.

The new PCB (see Fig. 28) is aligned by a couple of plastic tabs on the base. It is not screwed in like the old PCB was.

Figure 29. The new encoder cap has reflective elements built in.

Now might be a good time to examine the new encoder wheel and cap before we install them. Of specific interest, note that the new cap (Fig. 29) has reflective elements on it. In the new encoder, the cap is part of the optical path. This is different than the old encoder. Don't install the cap yet though, we need to install the wheel first.

Figure 30. New encoder wheel installation.

To install the new encoder wheel, place it on the shaft with the long side down (as directed in the US Digital instructions), then put the plastic spacer above the wheel. Using care, press firmly on the spacer such that the wheel slides down the shaft, and everything is pressed together and aligned correctly. Then remove the spacer.

Figure 31. New encoder wheel attached.

Fig. 31 shows the new encoder wheel attached. It looks nice.

To be continued...

 

[collinsmark]

Part IV (Continued): The New Encoder

The new encoder cap just snaps on (Fig. 32).

Figure 32. New encoder cap installed.

Figure 33. New gap in new encoder solution.

Fig. 33 shows the new gap due to the hole size in the new base being smaller than the cylindrical extrusion/lip in the motor housing. I was worried about this at first, but once everything was put together, it seems to hold itself in place pretty solidly.

Figure 34. Old electrical connector attached to new encoder.

The old connector attaches to the new encoder just fine. When I ordered my encoders from US Digital, I ordered new connectors too, just to be safe, in case it was the connectors that were causing the failure. But I didn't need to use one here, since the old connector seems to work just fine.

Figure 35. Fred's QA inspection of motor assembly.

Fred (Fig. 35) begrudgingly passed my handiwork as "acceptable." I was however cited for cleanliness, stating that my "overall hygiene" was of "serious concern."

Now, to prepare for re-installation, it's time to get one of those cable-ties ready. Figs. 36, 37 shows how the cable tie slides though a couple of openings in the thingy that holds the choke coil.

Figure 36. New cable tie. The figure shows the cable tie as it passes through one of the two openings, but it ultimately needs to go through both openings before it wraps around.

Figure 37. Getting the new cable tie and choke coil ready. Don't tighten the cable tie yet! The assembly needs to be reinstalled on the telescope before the cable tie is tightened.

To be continued...

 

[collinsmark]

Part V: Reassembly

Back out to the telescope.

First thing is to make sure you have the spring (the important one, Fig. 38) and those two larger hex screws.

Figure 38. Don't forget about that spring before you screw in the motor assembly via the two larger hex screws.

Attach the motor assembly in more-or-less the the reverse order of how it was disassembled. Just make sure that spring is in place before tightening the two larger hex screws.

Figure 39. Reassembly.

I actually used a red headlamp for reassembly. Don't forget the electrical connections (these are easy, since none of the connectors contain the same number of pins) and the declination adjustment knob.

Figure 40. Tighten the cable tie after motor assembly is reinstalled.

The final step (not shown) is to first re-remove the clutch knob and cylinder thing. Be careful though! Remember, that cylinder thing can slide off and the big wheel gear is not secure. Then reattach the housing cover, and finally, carefully put the declination clutch knob and cylinder thing back on one final time.

Figure 41. Success!

It's been a few weeks now since the procedure, and I've tested it most every night (at least a little) before the clouds roll in. Everything is in working order, as good as new.