AM5 erratic DEC backlash

Fredvallee

Hi, I have a AM5 bought early december 23 (288s PE period). Since the beginning I noticed that there is some backlash on the DEC axis, a very disturbing one indeed. Attached are two reports of the guiding assistant of PHD2. As you can see, on one of them, the backlash looks like a classical backlash coming from the drive belt, BUT on the other one it is very strange, is is a delayed backlash after the reversal.
Did anybody experienced the same kind of behaviour? Is it normal? Should I complain or not? The price tag is a range where you expect something working appropriately and I am frustrated. I struggle to achieve a reasonable guiding performance. I am in the range of 0.5-0.8arcsec (forgeting the jerky time window when it goes up 1.2arsec) with an evoguide 50/242 + flattener, Altair GP3 290M (ASI290MM impossible to source), PHD2 Ascom driver to get 12bit depth (with the direct driver offering only 8bit depth it is a nighmare). I am guiding an APO 90/f6.
Thank you for your insight.
Frédéric

Kevin_A

You may have a loose Dec belt. Check that first as the only source of backlash can be the belt being too loose or too tight. It should have a slight deflection of 1.5-2mm when pressed lightly with a finger or tool.

Fredvallee

Hi Kevin, thank you for your advice. I already opened it just to see and is seems to me that it is the opposite: the dec belt is extremely tight, I mean there is no deflection at all if I push smoothly. I was even wondering if it was not too tight. Besides this I do not know how to tune it. Is there any ZWO operation mode to do it. I do not want to send it back if I can do it mysefl.
I bought it to a european supplier but now I am in Morocco with it.
Frédéric

Kevin_A

Fredvallee yes, an overly tight belt can cause this too and if cold outside it can get worse.
There are 2 pdfs available for belt adjustment. Each stepper motor has 4 shc screws holding it down and 1 tapered screw for adding more tension or when loosened while the 4 socket head cap screws are loose, it will allow you to reduce the belt tension. Belt tension is correct when you press it down lightly and it deflects 1-1.5mm. Here are the files you need.

am5-dec-belt-adjustment.pdf

708kB

am5-ra-belt-adjustment.pdf

565kB

MrAstro

Kevin_A
Thanks for the attached files..
In the pdf files there is no mentioned about the "tapered screw" you said it will add more tension !
Should I use that tapered one OR just leave it ?

Clear Sky!

Fredvallee

Hi Kevin,
thank you for the docs. I used them to retight the dec belt and it improved the behaviour. The belt still has some flexure when you smoothly push it and I now have a consistent backlash curve showing _700ms backlash (divided by at least 2). Is it reasonable or have you tuned it to get even less? how much?
Thanks again, Frédéric

Kevin_A

MrAstro the tapered head screw is a positioned that really just creates a hard stop. You can use it to fine tune but as stated above… tension by hand and do not have it too tight.

Fredvallee

HI MrAstro,
having done it I can tell you that: there are only 4 allen screws fixing the DEC (as well as RA) motor. The last one (the one at extreme right in page 5) is intended to help you to tune the tension of the belt. It is a conical head one and according to the way you screw or unscrew it, it is more or less a fixed point to secure the tension. In my case this screw was already screwed up to the point that the full head is below the aluminium case board edge of the DEC motor. The only way to get the adequate tension was to push on the motor itself to add some tension on the belt and then try to tight the 4 allen fixing screws without releasing the tension. It is a trial and error process but you are converging fast. You need to check that the resulting tension on the belt is ok, let's say around 1mm. I used a large plastic shape to press on the belt to see how easy it is to move it. In my case I most probably will dot it again as the resulting backlash measured with the guiding assistant of PHD2 is now much more consistent in the way it shows up but is still high in my opinion at 700ms.
CS Frédéric

Kevin_A

Fredvallee awesome! Just remember that phd2 sometimes shows erroneous measurements for backlash. If your belt is perfect tightness… there can be no backlash due to the belt. 700ms is 3/4 of a second and if you visually slew with the cover off and the belt in full view it will look instantaneous and that cannot be 700ms. Give it a try and then go out and enjoy imaging.

Fredvallee

Hi Bruce,
I will check with the cover off. According to your experience does it mean that one should not stritcly apply the recommandation of the guiding assistant regarding backlash setting? I was already conservative with a lower backlash than the one proposed. What is your advice?
Thanks Frédéric