Balancing your GEM
BALANCE PRECISION - How well should the system be balanced? Probably several thoughts here. My personal opinion is that it should be relatively close. But I don't think perfect balance is necessary. In fact, it is near impossible. Doing visual work and just changing an eyepiece, adding a barlow, rotating the diagonal, or changing focus, will upset the balance from the "perfect" state. Besides, I feel a mount such as the Titan should be tolerant of some unbalance. My opinion.
But it would be nice to have a basic understanding of how to balance well. It is up to the user to go to the degree that will satisfy. Understanding this, there is no doubt that the most accurate way, whatever that may be, would be done with the worm disengaged. That is so that it is not in contact with the gear. In this condition, the mount can rotate freely on its bearings without any drag induced by the friction of that large plastic washer that does impose some friction even when the mount is rotated with the clutch knob loose. So when the worm is engaged and the clutch knob is loose, the mount cannot rotate on its bearings with the motor off. So instead, it slips on the plastic washer. Things like surface finish, coefficient of friction of the materials, etc. effect how much drag is present. I choose to balance with the worm engaged. And I use a spring scale. It's a 4 lb scale with ounce graduations. See Fig.1

Fig.1
The scale must be used at the angle of your latitude. I find it's use to be quite accurate. If I place a 2oz. weight on the countershaft, I can see the difference from when it is not there. Also, I had the mount balanced from the day before and I was going to regrease the worms. When I disengaged the DEC, it did not move! I was tickled.
So now, let's get into balancing.....
The first axis to balance is the Dec. Rotate the Ra axis so that the counterweight shaft is level with the ground and snug the Ra clutch. This puts the Dec axis parallel to the ground which puts it in it's freest position. Now with the Dec worm engaged or disengaged, (your preference), you're ready to balance it. All things that are to be used for the planned telescope session must be attached. It would be smart to have your focus in the ballpark too since it involves moving a mirror, or the attachments by way of focusing. So now you can slide the assembly in the saddle to get it balanced. I don't like this but it works.

Fig.2
I'll continue this topic by showing the screw in the dovetail plate in Fig.2. It is against the saddle. That means the Scope assembly cannot slide to the right in the image. I just did this for safety. So I do not balance my Dec end by sliding the telescope in the saddle. It's heavy and seems so crude. And not easy to move just a small amount to dial in a balanced condition.

Fig. 3
In Fig.3 you can see a white colored disk. It is 2-part epoxy painted LEAD. It's attached to a Losmandy DA adapter. I have several sizes of these weights but this one seems to work for everything. It's molded in a coffee can and is about one inch thk. To balance the telescope assembly, that is with all hardware attached, I simply slide the DA adapter along the length of the tube. This is very easy to do. The adapter has a "T" bolt to tighten it in any location. The "T" bolt can just be seen at the bottom edge of the disk. The disk is attached to the adapter with a stud and wing nut. It works great. It's shown in a forward position. My camera, filter wheel, everything, is balanced from this single item. A side benefit is that it also acts as a stop. Just like the screw on the other end. I have a second adapter that I did have right against the saddle. Don't know why it isn't there now. I can see it in the background.
It seems that balancing the Dec with the worm engaged or not, sliding the assembly in the saddle or sliding an alternative weight like I do should get the job done. This is not necessarily true. It has to do with the center of gravity of the assembly. Lets say you have balanced to perfection. Now if you rotate in Ra, does the Dec stay where it was? Probably not. But it should to be really balanced. After balancing, you should be able to rotate any axis to any position without the Dec moving on its own. If it does, it's due to the location of the center of gravity. Look at Fig. 4 It shows the balancing position

Fig.4
In the Top view you can see a green circle with a black vertical line through it. That represents the location of the center of gravity. It is shown on a white line representing the Dec axis. It is on the Dec axis because we balanced the Dec and by doing so, we put the C/G on the axis. We just didn't know it. But look at the Front and Right side views. In those views, the C/G IS NOT on the Dec axis. We didn't know this either. Now rotate in Ra and the image looks like Fig.5

Fig.5
In Fig.5 we see that the C/G is now ABOVE the Dec axis. It doesn't want to stay there. Gravity wants to bring it lower so the whole OTA assembly wants to rotate which indicates that we were NOT balanced. Again, a really balanced state is one in which we can rotate the Ra and/or Dec into any position and nothing would move.
Okay, how do we do that? Perhaps by carefully planned placement of our attachments. See Fig.6

Fig.6
Fig. 6 is a view looking at the end of my 12" Meade. My finderscope is at top right. Notice where I moved my Robofocus. It's the blue thing behind the wires, and that is not where it used to be. But in its new location the C/G has been moved to the Dec axis (yellow line) or at least closer to it. This alone has made a major improvement in my ability to rotate in any axis and not have anything move.

Fig. 7
See Fig.7 = The SBIG Adaptive Optics is mounted to the Optec focuser. The camera and filter wheel are mounted to the AO, which places them off from the OTA centerline. That is a lot of offset weight which throws the center of gravity way off. So I added a shortened ScopeStuff mounting rail to the Losmandy adapter so the weight could be slid sideways as well as fore and aft. In that image you can see that the weight is offset from the OTA centerline in the opposite direction of the camera. Fig. 8 shows a close-up...

Fig. 8
Now for the Ra axis.......
Again, worm engaged or not, is up to you. And again, more "near perfect balance" can be had with the worm disengaged.

Fig.9
There's more than meets the eye. First of all, all of that weight on the counterbalance shaft totals about 63lbs. That is what I need for my 12" Meade.Where should it go? The pieces can be spread apart into different groups and spacing and still be able to balance the telescope hardware. Well I have read on some forum that it should be choked up as far as possible (to the right in Fig.9). The idea being that it causes the least amount of shaft flex. This makes sense to me because the shaft is stiffer the shorter it's used length. If I moved the pair any more to the right, the single one would not have any more room on the left for adjustment. This is picky but probably smart for imaging.
Don't forget, the more accurately you get balanced at the start, the less you will depart from a balanced state when you rotate your diagonal or whatever.
Also check the section "Unbalance to Load the Worm"
Balancing Update - Sept. 24, 2004
As you may know, I dislike and do not remove or disengage my worm to balance. But I came up with something that works just like a disengaged worm but in a small way. That small amount of worm to gear clearance is the "small way". Here's what I do for each axis:
1- Rock the axis through it's clearance in one direction until the worm bottoms then let go.
2- Lightly push it in the same direction again. Does it go? If it does, the opposite end is heavy. Adjust weight.
3- If it doesn't go then push in the opposite direction and let go.
4- Lightly push it in the same direction again. Does it go? If it does, the opposite end is heavy. Adjust weight.
5- Repeat this with weight adjustment so that when pushed in one direction it stays there and when pushed in the opposite direction it stays there as well.
I have done this and then removed the worm block to see if it truly is balanced. Well of course it is! This is so cool. Give it a try. It works in any axis orientation.