AJAK H-H Magnet Wheel Modification

[linuxman]

While waiting for the 1 inch of snow we got to melt away and for it to dry off outside, I brought the AJAK H-H 180 mount inside for deep cleaning and to take a close look at the magnet wheel.

I have shot some pictures below for our observation and input.

The wheel itself is 2.5 inches diameter and looks like it rides on a 3/16 inch diameter shaft.

Five magnets that yields 1115 counts from limit switch to limit switch. The counts are take with a VBox V that I use for tuning dishes. With the distance between magnets, I am sure my GBox which will actually be running this dish will double those counts. I also believe that the dish will start and stop on each of the extra counts. We'll know in a couple of days.

I have finished cleaning and putting all new grease on it and will soon have it ready for the pole, but will post those pictures in my other thread.

Take a look at the pics below, and let's begin discussing how we can modify the magnet wheel itself or add a sensor to get more counts out of this mount.

 

[phlatwound]

What's the diameter of the existing magnets, 1/4"-3/8"?

Unless you take the wheel outside of that housing the only thing I can think of (besides adding a sensor) is to add additional magnets between the existing ones, but they might have to be smaller in diameter.

 

[linuxman]

According to the tape measure in picture 2 above, the existing magnets appear to be 7/16 of an inch.

I think the wheel could be pulled out of the housing to work on it, just don't quite know how yet.

I'll wait on Anole to get his apart for any real experimentation and dis-assembly.

 

[phlatwound]

To clarify what I meant by taking the wheel out of the housing....wasn't sure what is near (around or against) that cavity where the wheel is, but could you extend the shaft beyond the cavity, and put a larger wheel, with more magnets, on it, and still cover it up with the big (plastic?) outer cover?

 

[Anole]

quick thoughts:

I'm working with much smaller magnets.
Retrofitting that wheel with 10 instead of 5 should be a no-brainer.
But, let's see where we can go with this idea.
What would be a reasonable number?
Or, what will the Gbox handle ?
Is it worth doing 12? 16?
How many would give a similar count as your Birdview with 24?
Probably less than 24 magnets, as this motor has different gearing.

I have some tiny reed switches to go with my little magnets.
The reed that's on the unit will be the first thing to go!

What holds that disc onto the little shaft?
Do you know what's behind the disc?
I guess I can look at my own.
Gotta get the motor off the pole. It's stuck.

 

[B.J.]

It's been a while since I've had the reed switch disk off my mount, so I'm not positive that the magnets are the same or not. I don't think I'm adventuresome enough to mess with the magnets, but I did try to change the reed switch disk on mine. That disk was nothing more than a piece of cardboard type material with the reed switch imbedded and glued on if I remember right, and the cardboard got saturated with water, and the water got inside, and the grease in there actually tended to trap in in there. I don't think the water hurt anything, but I was starting to have problems with repeatability, so I thought that I would get a replacement disk. SkyVision had replacement disks, but I was dissappointed to find that they were basically home made things, just a plexiglas disk with a reed switch glued on, and I'm not convinced that the spacing from the reed switches to the magnet wheel is the same. I'm now having a problem with missed counts, so I'm thinking about putting the old disk back on, but first I think I'll coat it with a thick layer of urethane or something to water-proof it.
But anyway, I don't think I would mess with the magnets. I'm not sure how the vbox works, but with my Drake, I can bump the motor in small increments that are less than what would result in a count, and the lack of repeatability I'm seeing wouldn't be helped by having more counts, since I'm pretty sure that it's either lost counts or free play in the gears. So I really can't see how going to the trouble that increasing the number of magnets would entail would be worth the effort.
But the real reason I posted was to warn against spending money on the SkyVision replacement reed switch, as it can be made better with parts that cost a fraction of what they charge.

 

[Cadsulfide]

Once you get the magnet wheel off and peek inside, it may be better to put a 2nd reed switch in parallel and mounted so that you double your counts without modifying the wheel.

 

[1captain]

Well I am a big believer in high counts Linux, but also think if it not broke, no need to fix it.

now I have a Ajax mount on a 10-ft CM dish an it performs flawless, it has been up an running for about a year now, with the powertech G-box V3000. an have not had to reprogram it due to lousing counts, or don't need to bump the dish east or west on ku satellites. an it doesn't matter if going west to east or east to west.

now just tuned in ABC news on G-28 ku, position #144 on G-box, an can go to 153 an still have signal but 154 an lost signal also the same #s going east, at 10 #s on the G-box signal is lost, but still gives you a fairly wide tunning area on ku, but will vary on size of dish used. now on this dish a Pan-9000 is what runs the g-box an with pansats you can not control the amount you want the dish to move like with a Diamond, Coolsat, Satopia an other receivers that I have put on the g-box. so the Ajax has 9 counts either west or east of its stored position.

now also just check same transponder on my 1.2m dish that I fabed up a gear step up driving runing a 22-magnet wheel, an on this 1 position #130 an if I go to 156 lost signal but have week signal on 155. so with this mount I have 25 counts to the east or west of you're stored satellite location. but this is a high count H to H mount from due south to east or west limits little over 2200, with the G-box.

now after using this Ajax mount, for quite a wile I would not mess with its counts, works to good. but if you are determined to add more counts I would do it like my home made 1 here with gears an with 2 screws to atach a gear to the magnet wheel in the Ajax motor, an now add a idler gear with a magnet wheel atached to it out side of the Ajax motor housing, with some kind of bracket to hold all. an this way you can put it back to stock just by removing the 2 small screws attached that you atached on the Ajax magnet wheel.

Good luck Fred, this is just my 2 cents on this here, but I will be watching this tread to see what you wind up with in the long run. should be interesting.

 

[B.J.]

...

Five magnets that yields 1115 counts from limit switch to limit switch. The counts are take with a VBox V that I use for tuning dishes. ....

....
now I have a Ajax mount on a 10-ft CM dish an it performs flawless, it has been up an running for about a year now, with the powertech G-box V3000. ....
.......
now just tuned in ABC news on G-28 ku, position #144 on G-box, an can go to 153 an still have signal but 154 an lost signal also the same #s going east, at 10 #s on the G-box signal is lost, but still gives you a fairly wide tunning area on ku, but will vary on size of dish used. now on this dish a Pan-9000 is what runs the g-box an with pansats you can not control the amount you want the dish to move like with a Diamond, Coolsat, Satopia an other receivers that I have put on the g-box. so the Ajax has 9 counts either west or east of its stored position.

now also just check same transponder on my 1.2m dish that I fabed up a gear step up driving runing a 22-magnet wheel, an on this 1 position #130 an if I go to 156 lost signal but have week signal on 155. so with this mount I have 25 counts to the east or west of you're stored satellite location. but this is a high count H to H mount from due south to east or west limits little over 2200, with the G-box.
.....

I mis-read this... thought you were saying that the Ajax had 2200 counts from south to horizon, but I see that that's from a different mount. But while being confused (98% of the time for me), I looked up my spreadsheet with the counts for the various sats, and see that I have something in the range of 660 or so from south to west limit. That was close to what Linuxman said, in that that would mean about 1320 counts, H-H, but then I thought..... 1320*5/6=~1100.... which made me think..... maybe I have 6 magnets rather than 5. So I went out to my dish (snow FINALLY melted out there anyway). Turns out no, I have the 5 wheel thing too. Took a picture while out there:

http://wejones.ftdata.com/mag-reed-c.jpg

This picture shows the CHEAP home-made plexeglass reed switch disk that Skyvision sells for $29.95 . Darn thing has about $3 worth of materials, and would take me all of about 5 minutes to make one, so that's quite a markup.

HOWEVER... while looking at it, I noticed that the darn disk fits so poorly, that it slipped out from under the upper left washer, and almost fell out! The reed switch is probably a good 1/8" or more further from the magnets than it should be, particularly because the original was thinner material. Anyway, I'm thinking that this may be the reason for all the lost counts I've been experiencing for the past couple years since I replaced the original disk.

Also, looking at the counts/deg rotation, I'm seeing aproximately 10 counts per degree, ie 1/10 deg per count. I can't imagine this not being good enough, assuming that the mover has a capability of bumping the motor for fine adjustment. I've seen the sats move around by more than 1/10 deg.

 

[Lone Cloud]

I agree with the sentiment expressed that you add one or more additional reed switches rather than add magnets, although trying it the way it is first would be what I'd do.

 

[linuxman]

Got the new scanning dish on the pole yesterday and have it partially tuned in.

I did move the limit switches on both sides to allow for maximum movement of the dish primarily for testing lnbs etc. It will actually stand the dish up further than the Birdview mounts which I didn't think was possible.

Taking into account moving the limit switches on both ends, using the GBox V3000 positioner, the total counts from horizon to horizon no stands at 2816 vs. 1115 with the VBox V.

As I had thought the GBox reads a count off of each end of the magnet, and will count them with a single click including single click movement.

I read about 35 counts between 135W and 137W near the end of the arc, so there is very little reason to modify this mount or it's magnet configuration unless you simply want super accuracy.

You just need to buy a GBox and double you counts.