How much arc would I gain (approx) ?

[Vipersan]

So ..to complicate the situation ..
I've just marked and removed the circular impregnated rotating magnate ..
and connected a meter to the reed ..
Rotated the magnate by hand in close proximity to the reed switch ..
It transpires that there are 6 magnatic points on the disc ..not as I thought just one..
shown by the yellow dots ..
Question ..does the QARL have only 4 or perhaps 3 ..
I'll try and investigate at some point ..but in fact this might throw a spanner in the works ..unless the magnates can simply be swapped out ..say a 6 point for a 4 ..
answers on a postcard to XXXXXXX
So in the case of the superjack TD24 ..6 counts per full revolution of the disc magnate ..
Hmmmm....

 

[Captain Jack]

So in this case, it wouldn't help having two sensors anyway due to the points being opposite one another so they'll be hitting both sensors at the same time.

I would expect the heavy duty version to have the same wheel but perhaps due to it being heavy duty, some cogs would be larger hence not allowing for precision. Unless you found a wheel with more magnetic points... could be an easy way to "upgrade".

 

[Vipersan]

So in this case, it wouldn't help having two sensors anyway due to the points being opposite one another so they'll be hitting both sensors at the same time.

I would expect the heavy duty version to have the same wheel but perhaps due to it being heavy duty, some cogs would be larger hence not allowing for precision. Unless you found a wheel with more magnetic points... could be an easy way to "upgrade".

..Exactly right CJ ...
Its easy enough to borrow the wheel from the QARL ..just to check ..
If it has six points ...then nothing to be gained ..
If 4 then I could substitute the 6 ...
Pity they dont sell upgrade wheel magnates ...or do they ??
lol

Might be possible to place the second sensor offset so it activates in the gaps between the original wheels magnetic points ...but those points may be too close together ..
But it would double the count

 

[Llew]

The QARL has the same count per full magnet rotation as your other one VS - six. Just checked on mine, using the same method as you (mine's still waiting to be installed on a polar mount to replace my existing SMW one, but I can't do that until I move the pole back so the LNB arm doesn't protrude over neighbour's garden).

 

[Vipersan]

The QARL has the same count per full magnet rotation as your other one VS - six. Just checked on mine, using the same method as you (mine's still waiting to be installed on a polar mount to replace my existing SMW one, but I can't do that until I move the pole back so the LNB arm doesn't protrude over neighbour's garden).

Cheers Llew ..
Saved me a job ..
I guess it's down to the gearing then as CJ suggested..
The only way to increase the pulse count in practical terms would be to fit it offset something like this photo..
..and yes it would double the pulse count ..so not sure it's practical.
rgds
VS

 

[RimaNTSS]

I think, even if you find the way how to double the pulse count, positioner can only count from 0 to 1000.

 

[Vipersan]

Ok ..I dont like being beaten so may well have a go at making my own disc magnet ..
I'll use an original disk to make a mold and cast a disk in liquid plastic ..
Then drill 8 positions holes of 1mm x 0.5mm around the perimeter face all equidistant ..
..as I just found that tiny Neodymium magnetic discs of this size are available on ebay in packs of 50 or 100
epoxied into the holes in the cast disc this should in theory increase my pulse count by 2 per rotation.
A project I can work on without diassembling the actuator.
You never know till you try.
rgds
VS

ebay item :- 140783537418
but they come in different sizes.

 

[Vipersan]

I think, even if you find the way how to double the pulse count, positioner can only count from 0 to 1000.

Definately _not_ true Rima..
The Vboxes count from zero to 999 then wrap around but the count continues at least twice that I know of ..

There must be a high bit register within the Vboxs software to effectively increase the count beyond 1000 ..but it simply isn't displayed.

 

[Llew]

How about a simple pulse doubling circuit in line between the actuator and the V-box VS?

If you can source a voltage to power it, the ubiquitous 4011 IC would give you a couple of gates to double the input.

 

[Vipersan]

How about a simple pulse doubling circuit in line between the actuator and the V-box VS?

If you can source a voltage to power it, the ubiquitous 4001 IC would give you a couple of gates to double the input.

I had considered this Llew ..and it wouldn't gain anything really as doubling the pulses electronically wouldn't produce a pulse that actually represented a 'real' position in relation to the rotating magnet ..
ie ..it would simply produce 2 pulses at the point where the reed is activated ..
Yes you would count more pulses ..but the position data would be the same ..unless you got the second or 'non' pulse to land exactly in the gap between genuine pulses thus shutting off power to the motor in a gap ..
It would ..sort of work ..but guaranteed accuracy ? ...returning to a stored sat position .?
I have my doubts.
..Due to loading on the dish ..and rise and fall on the arc ...the speed varies slightly ..and an electronically generated second pulse in theory could end up overlapping a genuine pulse during dish movement ..errors creeping in over time..
rgds
VS

 

[Channel Hopper]

You could try fitting an optical sensor, one of them is bound to drop straight in to the back of your gearbox.

 

[Llew]

I had considered this Llew ..and it wouldn't gain anything really as doubling the pulses electronically wouldn't produce a pulse that actually represented a 'real' position in relation to the rotating magnet ..
ie ..it would simply produce 2 pulses at the point where the reed is activated ..
Yes you would count more pulses ..but the position data would be the same ..unless you got the second or 'non' pulse to land exactly in the gap between genuine pulses thus shutting off power to the motor in a gap ..
It would ..sort of work ..but guaranteed accuracy ? ...returning to a stored sat position .?
I have my doubts.
rgds
VS

OK VS, hadn't considered the finer points of the counting process. Back to the proverbial drawing board...

 

[Vipersan]

You could try fitting an optical sensor, one of them is bound to drop straight in to the back of your gearbox.

Agreed ..it is an option CH ...but then youd have to get a low voltage supply to the gearbox that was permanently powered ..as with Llews idea ..
And then replace the rotating magnet with a black/white reflecting surface ..or broken solid/space arangement ..
Plus of course extra electronics to convert to a form the Vbox could interpret.
Not impossible ..just complex.

 

[RimaNTSS]

Once I have got Cathrein original actuator and positioner. How many counts it makes? Maybe this can work for your need?

 

[Vipersan]

Once I have got Cathrein original actuator and positioner. How many counts it makes? Maybe this can work for your need?

Yes that's what I was saying when replying to CH ...
Broken/solid rotating disc ..and extra electronics which need a constant supply .
The photo you supplied shows only 4 'fan' blades ..thus a reduced count
We already have 6 ..and want to increase ..not decrease..so a new 'fan' disc would have to be made and substituted with more blades.

 

[RimaNTSS]

OK, I see. Do you think by doubling number of counts of actuator you can make whole system work 2x more precise? I think the more critical point will be play in mechanical joinings, and there will be (it actually always exist) difference of antenna position depending from which way you approach satellite from East or from West. And this difference usually more feeling at the ends of the arc.

 

[Vipersan]

I agree the ends of the arc are where errors often creep in.
In my case the rising side of the arc ie West is covered slowly with the same number of counts ..whereas the Eastern descent is covered more quickly with the same number of counts ..
Thus I can't position the dish as acurately to the extreme East..
The mount was never really designed to allow as much arc coverage as I'm pushing it through.
So the more counts per inch ..the more accurately I can stop it on the Eastern side ..and will take longer to cover one sat position on the Western side of the arc ..which is already adequate.
This problem is specific to my setup and not a general problem.
The TD24 had more pulses per inch and just about did the job...but the QARL has less counts ..and is less controlable at Extreme East

 

[Vipersan]

Tomorrow I'll actually count how many 'clicks' it takes to get from say 27West to 58 West ..
..and how many clicks to get from 30East to 57 East (the Vbox with the Qarl counts in 2s not 1s )
I think you'll be surprised at the difference.

Ok ..no need to wait I just did the count for my own satisfaction..
27 West to 58 West (Vbox counting in 2s ) So I only counted remote presses ie clicks ..
result 376 clicks
30 East to 57 East
result 145 clicks
So very non linear travel ..and the further East (extremes) the worse it gets
rgds
VS

 

[RimaNTSS]

I've put all my data in spread sheet. Can not find there linear sequence, I even think it is impossible.

 

[Captain Jack]

My gbox does display over 1000 counts. Also mine apparently doubles the count by counting the on and off states of the sensor. Not sure how beneficial that is though