Tracking Problems

[rfdesigner]

Hi

I have an old Fullerscope MkIV mount (very solid mount, 359 tooth bronze worm drives on RA and DEC, stepper motors direct onto worm, homebrew drive electronics)

.. and I've been pushing it somewhat, using for relatively long focal length deep sky imaging. I knew it wasn't the smoothest of mounts but hadn't been able to pin down the root cause.. well I think I've come to a conclusion as to what the problem really is. For a while I was blaming the lack of a proper bearing (just plain bearings with well chosen materials). So.. I've tried upgrading it to needle roller bearings on both axis. It's made it almost dangerous with the motors disengaged as it moves SO freely..

And the benefit to the tracking?... didley squat.

So I took a trace of several rotations of the RA drive, and I've ascertained what the PE is (+/-50 arc seconds peak, +/-30 arc seconds typically..) but over a period of 4 minutes per rotation, and with the direct drive steppers onto the worms, there are no smaller waves on the curves.

This however can all be taken care of with PEC.. indeed I've tried, but its never really cured the tracking.

I've analysed the data from last night and ascertained that whilst the PEC can get the PE down a good deal, I still end up with quite wild and woolly tracking, moving rapidly over a short time, and is different from cycle to cycle.

To my mind it's all pointing at the bronze worm wheel.

My current options are:

new worm set for the RA. (some hundred pounds I think)

precision shaft encoder for the RA (~£600)

A new mount (if I make the jump I'll be looking at a Mesu or similar.. but at nearly £5k I'm going to have to wait)

The question is.. can the old worm set be revived?.. i.e. if I were to lap the worm and wheel would I be likely to improve things? or is there anything else I can do?

Derek

[skybadger]

Hi Derek,

I, like you, have an old mount. Its an old AE of Luton class C. This used a 10" bronze worm wheel and I report a similar sort of PE function but ... it does seem regular and smoothish, so that I have been able to remove PE on a good night when the spirits are willing.

I did use valve lapping compound to lap the worm in though. The reason being I wanted to make sure that the SS2K drives I was putting on the mount could handle the load so I wanted to remove as much of the sticky points as possible.

Over the course of a rotation there is a clear signal of out-of centre of the wheel overlapped with multiple single-tooth excursions and a nasty bump I must grind out someday caused by dropping it on the bench.

I suggest that lapping it using a battery drill attached to the worm and some lapping compound won't make anything worse. Make sure to do it in both directions for a good while and to clean it out afterwards by hand with degreaser and toothbrush. This should only work with a hard worm (SS) and bronze wheel. I maybe wrong but with a soft worm I wouldn't expect as much if any benefit of lapping since the worm will wear as quickly as the wheel.

The other feature to look out for is sulphonated greases which contain sulphur which leaches the bronze and can cause pitting. If in doubt any polyolephin thickened grease should do, including Lithium and ER2.

I am now looking for a good thick grease to mop up any residual inconsistencies such as kilopoise or even RTD cutting grease. Its just thick enough and it stays put.

Is your worm covered BTW. This is supposed to make a big difference. Small spots of dust will show up. I keep promising to make myself some but keep on covering with a tarp at the end of the night.

Oh, and can you attach a PE trace ? Its always good to share.

regards

Mike

[rfdesigner]

How's this?

Top left is the PE, derived from Phdguide with guide pulses turned off.

Bottom left is the idealised PEC, fundamental only, no harmonics.

Top right is the effect of removing PEC

Bottom right is the guide frame to frame changes assuming 100% correction (this seems to tally with what I get in practice)

vertical scale is pixels, ~1.1arc sec per pixel.

Horizontal scale is samples, ~0.8sec per sample

The worm is stainless steel, (on RA, mild steel on DEC.. Booooo!)

Note the bottom right tace gives an RMS value of ~1.5arc seconds (equivelent to about 3.3arc seconds FWHM, just for guiding, the seeing is added to this) The DEC worked out at 0.5 arc seconds RMS or about 1.1 arc seconds FWHM + seeing

I have never had to worry about the motors on my mount.. they are quite high torque steppers and can achieve a meridian flip in less than a minute.

I'll look into the lapping.

Derek