Bah Humbug! Stupid dual axis controllers.

[Ajohn]

If it uses opto isolators there shouldn't be any problems. Also I had no idea if the motor drive chips are in the mount or the hand controller. From the photo's it looks like they are in the mount so switching too often shouldn't cause a problem in the hand controller.

I can't tell from the information if the switches pull low or pull high. If SW was working to minimum costs and using suitable parts they are likely to pull low - there will be some volts on the connection to the switches that are independent to each switch and the other side of all of them will have no volts on them and all will be joined together. If the sides of the switches that are all joined together do have volts on them then shoestring have been a bit naughty. The volts they provide may be higher than the battery voltage in the hand set when the batteries are low - that can blow up many chips. They can turn into thyristors - bang - smoke comes out, probably blowing a hole in the chip if there is enough power available. More parts would be needed if that is going on but what parts and if it will work depends some what on the chip in the hand controller that is connected to the switches.

Or maybe there is a cheaper option on the web that doesn't use an opto isolator and makes assumptions about voltage levels in the hand set.

John

-

[Quatermass]

I nvere read your sig but you seem to have added a guide port, if thats the case then you cant really blame the hardware if you add some home brew to it and it packs up. Uou should look at getting a Synscan kit for the EQ5 or better still getting an HEQ5 espcially for a 200 newt with astro imaging in mind.

The problem with a lot of home brew stuff is you cant ever be sure id its going to work, be reliable etc and for astro imaging unreliable and wonky is making things hard for yourself.

The first one with added guide port blew because it got left in the rain but the others 3 were all left as normal and run from the supplyed battery back. The fact that they blew with normal use has put me off them for good if your having to pay 100.00 for the whole lot your going to want something more reliable then that for sure.

Sent from my GT-S5670 using Tapatalk 2

[JamesF]

Looking at the hand controller, there seems to be very little that could go wrong that would render the unit completely inoperative. Most of the components look like they could be eight darlington pairs to drive the stepper motors and if any of those failed I'd expect the rest to carry on working. There are a few other components connected to the speed and N/S switches, but again I'd expect it to work in most cases if those failed. Other than that there seem to be a couple of diodes in the power line that might actually handle things if the power connector polarity is the wrong way around, but I'm far from certain about that.

Other than that there's just the microcontroller. Presumably it handles all the logic of stepping the motors at the correct speed and in the right direction. I guess if that's dead then you're up a creek without a paddle

What the microcontroller does surely can't be that hard to reproduce, and perhaps the darlington array (if that's what it is) couple be replaced with a few ICs. I wonder if it wouldn't be possible to replace the circuit board in the controller with an arduino and a circuit board with driver ICs and a few discrete components for £30 or thereabouts. Plus writing the code for the arduino, obviously...

James

[Ajohn]

It does drive the steppers directly James? I would have expected them to use stepper drive chips if it does but couldn't see any signs of them in the shoestring photo's.

If they are using some sort of ell cheapo drive chip and no heat sink too fast a pulse rate is probably the problem. With luck the chip may have a real part number on it so replacements could be found. The best way to remove the old chip is to carefully snip all of the leads and then desolder each one carefully. Then solder in the new part. Desoldering wick will probably be needed to to remove the old solder. That's best if lightly tinned with solder 1st.

John

-

[JamesF]

As far as I can tell (and I would say that I have more of a "passing familiarity" with electronics than any real expertise) the microcontroller chip has some data output lines that drive the darlington pairs directly and the stepper motors are driven directly from the outputs of the darlington pairs. There are some diodes associated with each darlington pair too, but it's not clear to me what purpose they serve. I didn't take the circuit board right out of the box so I couldn't see all of the tracks and work out the exact arrangement.

James

[JamesF]

Oh, the microcontroller is an 8-bit Philips unit, but it is programmable so I assume just replacing it won't work. I believe the EQ3-2 and EQ5 controllers are almost identical. However the final gear ratios are different for the RA axis on the two mounts so I'd guess that the programming of the microcontroller is slightly different for the two, or they have different crystals, or there's some sort of additional control for the microcontroller that tells it to divide the clock rate by the right amount for the EQ3-2 or the EQ5. If I were designing the controller I'd probably take the latter approach with some sort of switch to pull an input low for one and high for the other and the hardware would therefore be identical for both. It would just need someone to make or break the switch before boxing it up and putting the right sticker on the back.

James

[Quatermass]

Or altenativly skywatcher could make them better so they last longer would be good. :D:D

Sent from my GT-S5670 using Tapatalk 2

[AstroTux]

If I were designing the controller I'd probably take the latter approach with some sort of switch to pull an input low for one and high for the other and the hardware would therefore be identical for both. It would just need someone to make or break the switch before boxing it up and putting the right sticker on the back.

I'd just use a standard jumper connection, but to be honest with you its probably easier and cheaper just to flash the microcontroller with what you want first and then put the sticker on the back .

I've been having a quick scan through the documentation from Shoestring Astronomy and it would appear that the ST-4 mod simply uses 4 wires and Ground to provide the switching ( just like a standard handset would probably do ). The interesting part is their modifications for other handsets. They use relays to provide electrical isolation instead of opto-isolators - that should rack up the project costs nicely :grin: .

As far as the internals of the EQ5 handset are concerned ( please bare in mind that I have never seen one so am looking at the low grade pictures in their eq-mod.pdf ) it would appear that Darlington arrays are not used but that there are 8 individual Darlington transistors to provide the drive for the motors. Would anyone like to check this?

Anyway, one good thing - the EQMOD website has a lot of rather nice information about gears ratios for a number mounts. Great news for anyone wanting to use my ATmega8 design controller instead :grin: :grin: ...........

[JamesF]

As far as the internals of the EQ5 handset are concerned ( please bare in mind that I have never seen one so am looking at the low grade pictures in their eq-mod.pdf ) it would appear that Darlington arrays are not used but that there are 8 individual Darlington transistors to provide the drive for the motors. Would anyone like to check this?

That's correct. Everything is built from discrete components other than the microcontroller.

James

[AstroTux]

That's correct. Everything is built from discrete components other than the microcontroller.

Thanks James, that's cleared that up......

[Ajohn]

I suppose they are through hole rather than surface mounted too. Old design maybe copied.

There is a cardinal sin with that sort of circuitry. Sometimes a pair of what ever they are are driven from the same pin on the processor. Fine but they take time to switch off so when going say low to high both are on for a small period of time and any old current can flow which can blow things up. Not always instantaneously either. Sometimes series resistors are added to limit the current that can flow but it's still bad practice given the driver chips that are available and have been for some time. The other way is to switch one of a pair off, wait a while and then switch the other one on using 2 lines.

Funnily enough I have the hand controller for an eq3 not far from me. The single axis one. Took some force to open due to the buttons. It uses an atmel uP that is flash programmed. They may well program that up to suit the controller type. Normally where there is some facility to reprogram firmware products are programmed the same way as it's a bit messy to program parts before they are on a board - especially surface mount parts. It's hard to find out what the "darlingtons" are. Marked H8550 and S8550. Trouble is one source said the H8550 was npn and the other pnp. They seem to have rather low vbe figures for darlingtons but when I worked on analogue electronics I never made use of any other than at higher power levels. The spec for these seems to be rather ambitious for the size too.

They do seem to switch the transistors individually so I'd guess that they fail because they are a bit weak for the job and maybe made more so by the use of an autoguide which will probably make them switch more often. Telescope load and voltages might alter the current they have to cope with as well.

John

-

[AstroTux]

It's hard to find out what the "darlingtons" are. Marked H8550 and S8550. Trouble is one source said the H8550 was npn and the other pnp.

These are not Darlington transistors but plain old fashioned epitaxial low power audio frequency PNPs; with H8050s being the complimentary NPNs. Think of a slightly beefier ( 800mA ) version of the good old BC184/BC214 pairing and that's what you've got.

Just as a quick point, talking to others it would appear that EQ5 motors are 50ohm across each coil of a unipolar motor. At 6 volts their maximum current draw would be 120mA ( 240mA at 12V ). This means that high power circuitry is not required to handle the motors. In a simple drive circuit each pole is simply turned on then off as required. When on the current would rise to 120mA and stay there.

In a more complex chopper design the current would be quickly ramped up to the rated maximum current of the motor and then either turned off, or held near that level through PWM to allow microstepping.

My guess would be that these circuits are of the simple type as they can be driven from low voltages. A typical chopper circuit would require a much higher input voltage to provide the initial rise to the rated max current of the motor.

[Ajohn]

Yes. That was my impression. They aren't darlingtons. A better replacement if they are failing on power dissipation would be some metal can types if suitable parts can be found these days. I've only worked on surface mount for a long long time now so pass. For plastic I would be inclined to look at the ztx751 etc as a replacement but from a decent manufacturer. RS Components would probably be the easiest to source for that sort of thing. They will take mail order off anybody. Farnel is more of a problem. I know from using them on automotive products that the ztx's are rather robust parts.

On types of stepper drive I would beg to differ but agree that transistor drive may provide more power but in real terms I have my doubt in this application. Some drive chips actually switch mode a sinusoidal waveform into the coils. The idea being to get more even rotation rather than steps. A quick search for low voltage stepper motor drive also bought up exactly what I would expect. MOS bridge with low on resistance and an internal charge pump to obtain the drive voltages for the MOS parts. The 1st one I looked at works from 2.5 to 9v and had a 1 amp drive capability. There are also mos bridges about. The sine drive type came up following a search for low noise stepper motor drive. Not sure if I really fancy quietening my vixen drive down though. Part of the noise problem is down to metal gears and worms+wheels anyway.

I would caution about just buying any old parts on the basis of it's part number. For one manufactures sometimes have their whoopsies and these parts have been known to get onto the amateur or ultra cheap market. The other problem is more recent. As an example I sometime do a bit of electronics at home and often need a voltage regulator. For these I decided to buy off RS. Looking I found a range of adjustable ones same part number different manufacturers and different prices. I bought the Chinese company made ones. The spec compared with the others was awful but fine for my purposes. Just have to hope that they do meet it.

I mentioned heavy telescopes by the way because I have no idea what goes on in overloaded stepper motors as the drives I am aware of always use far too many volts and a current limiting circuit. The aim in this case is to make them step as fast as possible. Where I don't know I mention it. One thing for sure is I will be looking to see what they have done in an alt az synscan mount as I may well put something on it that is heavier than they do. Also try to find out just how over loaded steppers behave. I'm always cautious when magnetics are involved. Excess load might even make them run backwards while being driven the other way.

John

-