I'll throw another vote in for stalled steppers.
As mentioned, increase the ramp up time(acceleration) and decrease max step rate.
Note: no experience with Bartelising a scope or using his software(except for a casual look at it a few years ago) but plenty stepper experience(CNC).
Also, have a look around for better steppers, most steppers can handle a few hundred RPM no load and much less when loaded, sadly people accept this as the norm, but there are newish chinese made steppers that get up to 10,000 RPM(!) no load, and still handle over a thousand RPM when loaded. Very popular with the CNC guys. Approx $50-$80 each depending where you get them, plus extra spent on beefing up your power supply.
I do have a question at this point. It sounds like you used friction rollers to drive the scope. Did you have any problems with microslippage? Or did the encoders fix that? It's the reports of microslippage that put me off the roller system but if encoders fix it, I would seriously consider it.
Thanks,
Rod.
When looking at installing a GOTO system for the 16" Tridob (though eventually I went with a commercial unit to avoid having to use a laptop), I did lots of research and a bit of experimenting of my own. Basically, a friction drive will always have some microslippage, especially when slewing rather than tracking, so you either need to have encoders attached to the axes to "update" the position of the scope or make sure the drive system uses some sort of geared arrangement to avoid pointing errors.
For example, the Servocat system uses a knurled wheel that presses a bit into the wood of the ground ring for the AZ and a cable wound twice around a roller for the ALT. It needs the encoders mounted on each axis to make it a "closed loop" system, with the Argo Navis or Sky Commander telling it where it is pointing and correcting for any slippage.
Precise Goto seems to need more accuracy than just OK tracking.
Last edited by tnott; 24-06-2008 at 09:14 PM.
Reason: typo
Thanks for your comments. I should update where my scope is at since this post was resurrected. I have nearly finished 'bartelising' it. Bojan says it looks ready to test under the stars but the weather has been terrible lately so I haven't had a chance.
I went with the epoxy gears solution. At this stage, no encoders are fitted but I have bought a pair and intend to install them when I have the rest of the system ready. Strictly speaking they are not necessary in a geared system as you say. I want them mainly for public star parties where the scope can easily get knocked out of position.
I have been following your articles about your tri-dob. It looks great! Could you explain a little more about the cable drive on the altitude. I have read about cable drives but can't visualise how they work.
You have me interested in your comments on stepper motors. Which brand are you refering to? I know a lot of people like Vexta steppers. Where can you get them at the price you mentioned?
Amusing the way fate steps in and solves problems for you. I just managed to cook the circuit board on my Bartels stepper controller system - not entirely sure how I managed that but for now, the scope can go back to sleep while I cool off a bit.
Peter
Sorry to hear that Peter. I hope not too much damage was done.
Hi Rod,
The steppers come under different names and code numbers from different vendors but seem to be of the same manufacturer or 'clones'.
The cheapest Aussie supplier I have seen is Oceancontrols with their MOT-122 for $79 being the fastest. Check their pdf file for motor comparisons(within their own range).
The cheapest I have seen anywhere is a UK vendor, Arc Euro Trade, with their 180Ncm motor (£18, around $40 aussie)being the most popular with CNC conversions. They have an even smaller motor that may be suitable for smaller scopes for only £10.
I'm sure the same thing may be cheaper elsewhere, in the US or on EBay for example(I don't use EBay anymore so don't know..).
Keep in mind that the holding torque quoted for stepper motor's is a pretty useless indication of a motor's suitability for scope driving if the scope has a worm and wheel drive, instead, go for the motor who's datasheet shows the fastest step rates and best dynamic torque. (ie, high current rating/low inductance) The high current rating will likely need a chopper stepper drive unless you want to use some massive resistors as heaters on those cold nights!
(eg: standard(linear) stepper drive with 40v supply and 5v 2A motors = 70Watt resistor minimum)
As an example, the steppers I use on my milling machine(and will be replacing in the not too distant future) are Step-Syn brand(Sanyo Denki) nema23's with a 117oz-in torque rating, but have such high inductance that they are very slow compared to newer motors of the same size and torque rating. I could possibly even use smaller motors.
I have been following your articles about your tri-dob. It looks great! Could you explain a little more about the cable drive on the altitude. I have read about cable drives but can't visualise how they work.
Regards,
Rod.
The driven altitude fin has a 1.2 mm cable screwed onto one end which runs along the curved length then wraps around a roller twice. It then connects to a spring, which is connected to a small toggle lever.
When the lever is engaged the friction of the cable around the roller drives the scope (like a horse's reigns outside a saloon in a Western). When the lever is disengaged the cable slips so the scope can be moved manually. A notch is made in each end of the bearing (to stop it slipping off) and a channel made in the teflon it sits on.
The cable also winds from side to side a precise amount depending on the width of the cable, as well as the circumference roller and the alt bearing. Mine needed a custom one.
Attached are some photos of more "standard" scopes
Thanks MrB for stepper suggestions. Unfortunately , the Oceancontrols steppers are all bipolar. Mel's system is designed for unipolar motors.
Regards,
Rod.
Rod,
It is possible to use bipolar motors with Mel's software.
Of course, the driver electronics must be differently designed, it should have "bridge drivers" instead of single transistor switch for each motor terminal.
The bridge stage uses 4 transistors per coil, so basically you need twice as many transistors (plus two small ones as drivers/inverters, so this is 6 per coil, 12 per motor). This is not a problem to build, though.
However, I would not worry too much about all this.
Because there is always something better behind the corner.. .. and if you chase the perfection, you will always try to improve but never really use and enjoy what you already have
EDIT:
I just had a look at data sheets (Oceancontrols steppers). 6-leads motor is unipolar, so you do not have to modify Mel's driver box.
I think the Oceancontrols data sheet is a modified version of the motor manufacturers sheet, yes it does include the unipolar motors but if you check the part numbers, OC only stock the bipolar motors.
Yes bipolar motors will work if the hardware is modified, though with these high current motors a chopper(PWM) drive is preferred.
These are reasonably cheap now but need step/dir input, I can't remember if Mel's software generates step-phase or Step/Direction outputs?
True, not worth worrying about, unless you find current slew rates extremely painful
I think the Oceancontrols data sheet is a modified version of the motor manufacturers sheet, yes it does include the unipolar motors but if you check the part numbers, OC only stock the bipolar motors.
Yes bipolar motors will work if the hardware is modified, though with these high current motors a chopper(PWM) drive is preferred.
These are reasonably cheap now but need step/dir input, I can't remember if Mel's software generates step-phase or Step/Direction outputs?
True, not worth worrying about, unless you find current slew rates extremely painful
Hmm. you seem to be right re data sheets.
Mel's software does not generate step/dir signals (actually it does but only for field rotator). However, it does generate PWM signals, just like choppers (without current feedback though), for microstepping, and the max value for motor current can be adjusted via microstepping parameters menu.
Any Bartelised scopes out there?
There may be plenty of tricks and solutions to be shared....
I just found this thread.
I Bartelised my GEM (10"f5 newt.) about 15 months ago now. I had to beef the voltage to 24 volt (12volt steppers) to get decent slew speeds. But other than than that, the scope works fine.
My final drive is by roller and I suppose it slips a little but it's no big deal as I make sure I sync on a nearby bright star before slewing to my target, especially after a polar flip.
I'm very happy with the rig.
There are pictures of my scope all over IceInSpace but to make it easy here are some again.
Dave, this mount looks really impressive!
You must have an access to a very decent workshop, or you must have a good friend (like I do but mine does not have as much free time and interest in those things as I would like him to :-) )
Mel's software does not generate step/dir signals (actually it does but only for field rotator). However, it does generate PWM signals, just like choppers (without current feedback though), for microstepping, and the max value for motor current can be adjusted via microstepping parameters menu.
Bewdiful, all you need then is 4 more transistors(and diodes) to modify for H-bridge drive and a higher voltage supply
Hello Rod,
well, I'm back and the new PCB has arrived from Mel Bartels, along with a few of the MC3479 field rotator chips (the ones I blew when I cooked the previous board). It looks as though the old board is OK with a new chip swapped in. Sadly I also cooked the printer port on the old Compaq I was using. I am definitely older and wiser. I realize now that I disobeyed one of the key instructions in the Manual - I hadn't connected up the motors to the PCB etc before powering up the board.
I have an AstroFest this weekend (in Qlod) and had hoped to finally give this big 16" scope it's first light there. But chances of getting it all set up in time are diminishing - so the forecast of continuing showers is almost a relief.
Peter
Welcome back. Good to hear your board is going OK now. Too bad about the parallel port. Do you need an old laptop? If so PM me. I have one spare I don't need.
My system is nearly ready too. I have learnt that it is something you can't rush.
Thanks Rod,
and thanks for the offer of a laptop. I'm OK there, thanks to a couple of grandsons who go through laptops like they do sox. I get their cast-offs that are generally quite useable.
Now that I'm back in action, I am still at the very beginning of the learning curve with Bartels system I'm afraid and in need of a bit of help with something seemingly very simple but ...
In the setup, in defining my local site, I have enterer Lat and Long for my site here in Qld as -27 Lat and 153 Long but for some reason, the system seems to think I'm somewhere in the northern hemisphere because any goto commands generally see the scope move towards a nadir position.
Is there some trick to entering Lat and Long for a southern hemisphere location?
You are a little ahead of me on this. I have actually been thinking about the same problem. Andy Martyn used to have a great site about his Bartelised scope. He had the following info on it which I hope helps:
"A note for those of us in Australia that the time zone should be a negative number to get the sidereal time right and that the longitude should be positive. Thus for my location in Melbourne at 145.06855E 37.81029S the values are:
Longitude 214.93 145, Latitude -37.81O29. TZ -10."
Does that answer your question?
If not, I'm sure Bojan would be able to advise you.