[Wavytone]

From an engineering standpoint, the clams by McCreary concerning accuracy are frankly very overstated, ie. b/s. Sure it is definitely better than no tracking, and in the era when McCreary came up with the idea, it was better than the crappy 1980's Celestron which had terrible period errors in the gear train and a pathetic RC oscillator controlling it which resulted in tracking best described as a "drunkards walk".

But seconds of arc ? Nope, no way.

It does make sense - in theory - to use a constant pressure based on a constant weight applied to a constant area on a bag full of water, with a simple needle value to achieve a constant rate of flow. But modern gears, digital electronics can do significantly better - if you buy a decent mount. Throw in closed-loop feedback - in the form of an auto guider, and you can have PERFECT tracking.

The McCReary mount assumes that while tracking:

- the force applied on the bag is constant,
- that the force is applied over a constant area.
- the viscosity of water is constant during the observing session,
- the properties of the needle valve don't change either,
- the axis of rotation is accurately aligned with the pole,
- ignores the effects of atmospheric refraction, and
- it also assumes the telescope on top does not flex as the angle of tilt changes.

However in reality all of these assumptions are not quite true:

1. the telescope tilts on the platform above. As there is no attempt to make this a balanced pendulum, this implies the pressure applied on the bag varies throughput the tracking range as degree of imbalance of the telescope is a function of tilt This means the pressure on the bag, and the flow rate through the needle valve are not constant, nor is the tracking rate as a result.

2. As the telescope tracks the angle of the plate pressing on the bag of water changes somewhat with respect to the horizontal. This matters, in respect of the pressure applied to the bag.

3. The viscosity of water changes with temperature, and this affects the flow rate through the needle valve. What this means is that as the night cools and the water temperature drips, the water flows more slowly as its viscosity increases. You can Google the Bernoulli equation for yourself.

4. No mount is perfectly aligned with the pole. For one of these , I'd be very surprised if it was aligned better than 1 degree from the pole. That means at most points in the sky it's going to drift slowly in both dec and RA, no matter how good the tracking rate.

5. Atmospheric refraction is a function of altitude above the horizon, its non-linear, and can amount to half a degree (1800 seconds of arc). That's a lot of seconds of arc. If your scope tracks perfectly at the zenith, it won't at lower altitudes, and vice versa. Given that it takes say 5 hours to g fro zenith to 15 degrees above the E or W horizon, this implies the tracking error accumulate around the order of 6 arcsec per minute. This is not inconsiderable, its non-linear and it reverses as the mount passes the zenith, to make matters worse. Hence the claims concerning accuracy are frankly rubbish.

6. Most dobsonians are frankly quite flimsy, and flex unpredictably. As for wood, it is quite flexible. For a visual dob being pushed around,, how it flexes is irrelevant. But when tracking, it does matter. In fact it matters a lot, as anyone with an equatorial mount can confirm.