Paramount MX - I think I might be the culprit

[baileys2611 (Simon Bailey)]

Hello fellow community members. I posted this in the Software Bisque forums but I know there's a few owners out there who might be able to offer some advice to me please?


Hi All,


Catchy discussion heading to make a light hearted attempt at what might be a serious issue.


I've done maintenance on my Paramount MX, taken the worm drives off and re-greased then put them back on. I think I need to replace the timing belts (the MX is about 4 years old) because I'm getting very little error on the DEC but RA is swinging between +1.3 and -1.5 error, giving me oblong stars over 2400 seconds.


So, I took off the cover yesterday and noted the belt's tension and cam shaft lock, undid the cam-shaft screw by about 1/4 turn, re-positioned it and re-locked it. All good. Then I looked at the grub screw above the cam-shaft locking screw and turned that about 1/4 counter-clockwise. Realised I was being an idiot and turned it back to where it was before.
Have I stuffed anything up? I tried to track afterwards and it could have been that it was a windy night (15-20km winds) but my tacking was terrible.


I'm going to take everything off, make sure there are no cables pinching anything, replace the belts and put it all back again but if I've stuffed something up by moving that small grub-screw then I might as well own up and figure out how to fix that first.


Any thoughts?



Thanks
Simon Bailey.

[gregbradley]

The original black belts can shed material and it can build up in the gear teeth. Did you clean all that build up off?

The cam lock pin is a touchy part of the PMX. That pin is very hard to get tight enough to actually lock the cam in my experience but it does not seem to be critical. The cam pin should be locked down but tends to pop up.

The screw above it - not sure what it does. It may lock the 2 vertical threaded rods/bolts above the cam pin. They determine the tightness of the worm and gear and too much and the motor will stall. Too little and the axis will slip on the gears too easily.

From memory those 2 top threaded rods are opened 2.25 to 2.5 turns from fully in.

The little grub screw's purpose is best asked of Software Bisque. Chances are its not connected with your tracking issue. I doubt it. Its probably locking something like those 2 vertical threaded rods I mentioned.

Tensioning the belt, making sure you got all the build up off the gears from the belts (I used a small allen key and got it in the teeth of the motor gear to get it all off) or poorly tensioned cam lock are likely causes.
The threaded bolts can also mess with tracking.

If too loose you can also get backlash and an oscillation back and forth in your tracking until things are at correct tension.

So, make sure the belt gears are clean with no tiny black rubber particles around, make sure the cam is tensioned well and if you are getting slipping of the RA axis under load then tighten those 2 vertical bolts until just right (firm but not stalling the motor and no slack when moving the RA axis with your hand).

Chris Venter has a couple of videos of replacing the belts and tensioning the cam.

Greg.

[baileys2611 (Simon Bailey)]

Hi Greg, thanks for replying.

I've got the grey belts, poly with kevlar, no black residue - however that is something I did when I had the drives off for re-greasing (as in cleaning the teeth).

Funny thing is that I've loosened the cam lock screw again since then and the pin does not pop up, as in I cannot push it down any further. However, I understand what you are saying about it tending to pop up from time to time and will keep an eye out for it.

Yes, the centroid of the star is oscillating left and right in my tracking image, it's not moving up or down very much (although the other night the wind was not helping me) but on good nights it's very stable in DEC and flopping around in RA (I think that's right - not in front of the latest log so I could have those reversed).

Thank you for describing what those grub screws are for. I'll have a go at those and see if it makes a good change.

Simon.

[gregbradley]

A simple solution is to try and put your counterweights offbalance a little and it probably will stop the oscillating. I had that happen a few times and that fixed it.

But it is a symptom of too much slack. A bit of backlash and these mounts don't really suffer from backlash. So perhaps redo the cam lock pin as per the Chris Venter video. Perhaps a slight adjust of those 2 threaded bolts. But remember where you started.

From memory winding them out loosens, winding them in tightens. But remember your starting point and do the same adjust to both ie. 1/4 turn inwards (clockwise) on both. I would check if there is any slight rock in your RA axis first (wriggling it).

Greg.

[baileys2611 (Simon Bailey)]

Hi Greg,

I've been discussing this on the SB forums also.

Your suggestion about slight weight imbalance, I've heard that before and will give it a try. I'll also try re-doing both worm blocks to see if that helps including tightening the grub screws a tiny amount.

The SB forum also suggested some belt conditioner (grip 'enhancer') and making sure I don't have too much grease on the drives as that will make things slip also.

The advice I'm getting all round is great I have to say. Thank you.

[PRejto (Peter)]

Hello Simon,

Chis Venter left out a very important part of reinstalling the worm. The 1/4-20 bolts that seat the worm into the ring gear. Not his fault as SB added the step after he made the video. If you didn't do the adjustment all bets are off and that easily could be the source of your problem.

What you describe as swings sounds to me like uncorrected PE. Are you seeing these 3 arc-sec swings with PEC on? If PEC is off 3 arc-sec Peak to peak is not at all a problem and you will get rid of it with PEC. If you have trouble with TSX getting PEC going I highly recommend PEMpro. It got my MX down to around .6 arc-sec.

Peter

[baileys2611 (Simon Bailey)]

Hi Peter,

Yes, I tried with PEC on and off and I've tried using TSX PEC and PEMpro - I actually started to get similar results from PEMPro as I get from TSX, but I think what you call the 1/4-20 screws are most likely my problem. If those are the grub screws on top of the worm drives, I didn't touch those at all.

I'll give it another go over the weekend, make sure I haven't over-greased or over tightened or under tightened, condition the belts, re-tension and re-do the whole PEC as well as t-point and see how that works.

[PRejto (Peter)]

Simon,

The 1/4-20s are the hex nuts on either end of the worm. Basically you loosen both, then press up on the worm block (this pushes the worm down into the ring gear), then tighten both bolts simultaneously. You need 3 hands to do this (or better 2 people), but I did it many times by myself.

Peter

[baileys2611 (Simon Bailey)]

Okay, pulled it all apart yesterday and put it all back together again, this time adjusting the cam screws on top of the worm block. I noticed one of the belts has some damage on the side, the kevlar is fraying a little at one spot...so I've ordered four new belts (two for now and two for the future), but I got them from a supplier in the US, exactly the same belt but instead of $30 each + $90 (USD) shipping they are $4 each plus $15 shipping (USD).

In the meantime, I've used the belt conditioning spray and tensioned the belts correctly. I purchased new motor mounting screws from a local supplier (stainless steel of course).

So lots of maintenance done, put it all back, balanced it nicely. I'll try an even balance first and if I'm having issues I'll try a small imbalance.

Now what I need to do is practice my cloud busting skills so that I can do another t-point run and look at my PEC. Hopefully with all the good advice and proper adjustments I'll get a good result.

Thanks Greg and Peter.

[gregbradley]

There was a time when PMX belts were not the right quality and prone to fraying. I think the belts could easily cause a bad spike in PE if there was some damage in it.

Wow, its that how much they mark them up. Nice find.The $90 shipping is outrageous when its a tiny little belt that could fit in a postpak envelope.

Greg.

[baileys2611 (Simon Bailey)]

Yep - and it looks like I'll be needing a new belt.

Measured PE using Pempro and applied a curve, +1.5, -1.8 in the end which is pretty good, but the end result is still moving around by on average 1 pixel. I've gained some accuracy, which is a good result I think.

On the negative side the RA error has risen a little bit, but on the plus side at least the stars are less eggy and more round

I think when the new belts come I'll install them without removing the scope. That thing is heavy and needs safety gear to lift it properly

BTW - for any PMX owners out there I'm happy to order some spare belts and get some spare motor screws and send them around the place (for cost of postage and parts). Assuming that these belts come soon, it will have been a smooth transaction.

[PRejto (Peter)]

Hi Simon,

I'm a little surprised that you are not getting a better PEC. The mount is certainly capable of sub arc-sec peak to peak. Not sure how your seeing was but you might want to try again on a better night. Did you use PEMpro or TSX? I always got a good result from PEMpro and had real difficulties with TSX.

What is your uncorrected PE? Is the mount within spec?

Peter

[gregbradley]

I agree 3.3 arc secs is good but the mount is capable of better with a good PEC curve.

A good Pempro curve should make quite a difference. Also a good Tpoint model with the accurate polar alignment routine helps a lot.

Greg.

[baileys2611 (Simon Bailey)]

Sorry - I should have said that those are in pixels. My resolution is 0.65 arcsec/pixel so the arcsecond PEC is actually +0.975, -1.17.

I used PEMpro for this, I can get a pretty good result with it too and seeing was ok - but we tend to get a bit of haze in Murrumbateman, not the easiest place to do astro-imaging (I like a challenge).

Re uncorrected PE - how do I measure that?

[PRejto (Peter)]

Hi Simon,

First you turn off PEC corrections in TSX (Bisque TCS).

Collect PE data as normal (Protrack can be on. It really doesn't matter).

Run PEMpro for a few cycles and take a look at the curve. That is your uncorrected PE.

(To apply that data for correction, if desired, follow the PEMpro instructions on how to paste that curve into the Bisque TCS. You want to paste the curve into the same window that displayes the curve that was saved to the mount. There is an option to retreive the curve from the mount and to clear.

After you load to the mount enable PEC in the Bisque TCS and do the measurements again for a cycle or two. compute the error in PEMpro. Hopefully you then see a good correction. If the error has increased it means the curve was applied with the wrong phase. Invert the original curve and repaste it into the TCS.)

Sorry if I've repeated stuff you already know!

Peter

[baileys2611 (Simon Bailey)]

No problems Peter and thanks for the help.

I did an uncorrected run tonight (10 worm cycles) and was getting around +4 to -3 arcseconds in PEMpro, I think the mount specs are 7 arcseconds peak to peak. BUT, this is with a dodgy belt that I am waiting to replace.

After applying the curve, I did a second run in PEMpro with PEC applied and it got worse! I don't know how because the star was always centered in the 100 by 100 box, but I was getting reports from PEMpro of +15 and -10 arcseconds. That can't be right.

So I re-did it all again and got different results the second time, ranging from +10 to +1 uncorrected. Applied the curve and same results, worse.

Then I re-ran the calibration wizard and did a third run (10 worm cycles), +9 to -3 this time and exported the data, applied the curve a re-did the run and it was not good. Worse again. I re-analysed the data and inverted the curve and still got a worse result. I then adjusted for drift and re-applied.

I will try using TSX tomorrow or this weekend and see if I can get better results that way - but at this point I'm getting better results without PEC applied than with!

[PRejto (Peter)]

Hi Simon,

Well that s pretty odd! Your uncorrected PEC is just within spec. On my MX uncorrected was around 3 arc-sec. I never had a PEC from PEMpro not work. Certainly it won't work if inverted, but whenever I changed the phase things got a lot better. How Was your seeing? Late at night is usually better. By the way, Ray stated that one really ought not to do more than 5-6 cycles. He said doing more makes it harder actually for PEMpro to see the PE. So, you can save time next time you try. Personally I wouldn't expect TSX to do better. In fact only a few people I know got a good result with TSX in the Southern Hemisphere, and, nearly everone reports that PEMpro does a better job. Are you certain that all the settings in PEMpro are correct? I'm not sure if the MX and MX+ have different worm periods. They might so be sure you have the right mount selected!

Another option you might condsider is that you can try swapping your two worms. It might be that your DEC worm has a better spec than your RA worm. A fair amount of work, but I'd persist with things as they are. I take it that you did adjust the 1/4-20 bolts? Things are quite firm in RA? You can check this visually by pointing at a distant object during the day (could be with a high powered finder scope if you don't want to remove your camera). Using a reticle notice if when you rock the scope gently in RA that the object you centered returns to center after releasing pressure. It should and if it doesn't is a sign of hysteresis.

Peter

[gregbradley]

PEC has to be oriented correctly to the correct position of the scope. Did you go through all steps of the PEC wizards with Pempro? There is a mount setup wizard as well. If the scope is on the west side of the mount for example and the curve gets sent to the mount that is reversed it will make it worse.

Greg.

[baileys2611 (Simon Bailey)]

Originally Posted by PRejto

Hi Simon,

How Was your seeing? Late at night is usually better. By the way, Ray stated that one really ought not to do more than 5-6 cycles. He said doing more makes it harder actually for PEMpro to see the PE. So, you can save time next time you try. ... Are you certain that all the settings in PEMpro are correct? ...

Another option you might condsider is that you can try swapping your two worms. ...

I take it that you did adjust the 1/4-20 bolts? Things are quite firm in RA?

Peter

Seeing was good (except for the full moon) last night because the temperature dropped to 2 degrees (or thereabouts) and restricted any heat haze and dust.

Got it 5-6 cycles next time, no problems.

Settings in PEM pro - yes, I went through the wizards and completed them and selected MX (not MX+ or the other one...temporarily forgotten what it's called).

Swapping the worm drives!! What a great idea! I'll definitely give it a try because the RA is quite firm and tracks very well. That will isolate an obvious mechanical fault with the worm drive...but will that stuff up the encoding or is that not a problem?

I definitely adjusted the 1/4-20 bolts, turning each 1/4 of a turn at a time to get around not having three hands! I turned them all the way in, then back out 2 turns only.

Is it that the less you turn them the tighter the cam will be or is it the other way around?

[baileys2611 (Simon Bailey)]

Originally Posted by gregbradley

PEC has to be oriented correctly to the correct position of the scope. Did you go through all steps of the PEC wizards with Pempro? There is a mount setup wizard as well. If the scope is on the west side of the mount for example and the curve gets sent to the mount that is reversed it will make it worse.

Greg.

Hi Greg,

Yes - I went through all the steps of the wizards, including re-adding the mount through the mount wizard.

You know it's funny how thick I can be sometimes...All this time I've been reading 'scope on the west side of the meridian' to mean that the scope is pointing to a star on the west side of the meridian.

So, now I think I understand it to mean if the scope is physically on the west of the mount. It's always been on the east of the mount, pointing to a star that is on the west of the meridian but as close to 0degrees dec as I can get.

Is that the right set-up?