Are all these earth connections necessary

Thread Starter

faceter

Joined Dec 4, 2011
7
My project is control of my telescope via my pc and Ascom. I have completed this aalready but it failed - rubbish soldering I think. Not done much of that.

I am using an Adruino Uno via USB and the PC.
In turn the Uno is connected to my scope's hand controller via my failing Veroboard construction. I should add here that I am using another persons schematic which had worked for him.
Pls see the attached part schematic.
Question 1
Related to the connections to Gnd. I don't undersand why so many. Can a common Gnd be created rather than all the connections in the schematic? From what I see I am losing 50% of the use of the ILQ74 Optocooupler. I could do other things like control a Dew Heater for the main lens and eyepiece. Am I right, or is it necessary to ground the unused connections?
Question 2
From the same schematic it can be seen that the signal passes through the ILQ4 and goes to the various pins of an RJ12 socket. Now the question - hw to connect to the RJ12 socket. I've tried soldering and ended up with a rather strange black blob. I've tried some terminal ends but they fall off. Any pointers on this for me?

Great site.
Hope this is not too annoying.
faceter
 

thatoneguy

Joined Feb 19, 2009
6,359
Do you have a larger version of that schematic?

From the way it looks, the left side grounds are the return path for all the diodes in the optoisolator, so they are required.

For the jack, all the signals need some reference point, which is ground since differential signaling isn't used.
 

ErnieM

Joined Apr 24, 2011
8,377
If this really work for "another person" then he did not document what he really did.

Your picture schematic only shows one small part of the unit. I would first check if there is some limiting resistor between the Audino and the diodes, each line should have one.

Next, as drawn the transistors are not connected so they can do anything. I would expect the emitters to be grounded, and the collectors to be going to the connector, just the opposite of what is shown.
 

SgtWookie

Joined Jul 17, 2007
22,230
Here's a datasheet for the optocoupler:
http://www.datasheetcatalog.org/datasheet/vishay/83640.pdf

I took the liberty of re-drawing the schematic a bit larger; I couldn't read the colored signal names from it though:



There was an error in the original on the output side; the grounds and signal pins were swapped.
The input side didn't show resistors to limit the current through the IR emitters, which is not a good thing.

And finally, yes - you do need to use all of the grounds. Without them, there would not be a complete path for current flow.
 

Attachments

Thread Starter

faceter

Joined Dec 4, 2011
7
Hello
Thanks to yoour re-draw of the schematic I can see where some connections were wrong. My understanding of the Opto-Coupler has greatly improved.
I have attached a larger version of the scheme as Thatoneguy asked for and also for ErnieM.
I have re-read the original scheme and method statement.
There is no mention of any resistors on the input side of the ILQ74 and he certainly got the earthing side wrong. I can see that myself now.
I should not have blindly used the scheme without checking it out. I have spent a lot of time on this and learned much from the pitfalls thanks to you all.
I really appreciate your help.
There's only one more thing - the connection to the RJ12:( Anybody have a solution for me?

Thanks to you all
faceter
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Well, what I would do is get an RJ12 socket that was suitable for mounting on a circuit board, perhaps like this one:
http://uk.farnell.com/molex/43860-0002/modular-connector/dp/1654815
or like this:
http://uk.farnell.com/fci/87180-066lf/socket-right-angle-6way-6wire/dp/1097904
or this:
http://uk.farnell.com/fci/90512-001lf/socket-vertical-6way-6wire/dp/1097914
or this:
http://uk.farnell.com/mh-connectors/mhrjj66nfra/socket-rj12-6p6c/dp/3938359
...and mount it on the same board as the optocoupler, then wire it up, and use a standard 6-wire phone cord to connect the board to the telescope, or whatever it is you are connecting the RJ12 to.

By the way, you can get a 4N25 optocoupler:
http://uk.farnell.com/vishay-semiconductor/4n25/optocoupler-transistor-5300vrms/dp/1612453
which is just 1 channel, and use it for a 5th signal line to your telescope. You can connect the grounds together with those on the same input and output sides as the ILQ74 4-channel optocoupler.
 

Thread Starter

faceter

Joined Dec 4, 2011
7
Hello again and thanks for the reply

I have in fact bought a couple of RJ12 items which I hope will be OK.

http://uk.farnell.com/mh-connectors/mhrjj66nfras/socket-rj12/dp/3938372?Ntt=393-8372
http://www.rapidonline.com/Cables-C...430/?sid=5555b695-da5b-4f89-a22e-b897ad7cf0a0

You suggest placing these on the same board as the Opto-Coupler. How would that be done? I am using Veroboard for my project and the hole pitch doesn't match the pins on the RJ12's:confused:

I've attached a very brief outline of what my project is about as you weren't sure where I was going to stick the wires. I hpe this makes it clearer. Any suggestions re a 'smarter way of doling it' keeping in mind my very amateur status would be appreciated. My work tend to look uglythough I am trying to improve my soldering skills.

Thanks again.
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Those look like they will work.

You suggest placing these on the same board as the Opto-Coupler. How would that be done? I am using Veroboard for my project and the hole pitch doesn't match the pins on the RJ12's:confused:
Ahh. Yes, that's a bit of a sticky wicket. Usually, one would make a custom circuit board for that kind of a jack. Since you're using Veroboard, you will need to drill or cut a hole in the board for the pins to go through, and then you will need to solder on jumper wires. If you have some spare CAT-5 cable or telephone wire lying about (not the cables; the wire that goes in the wall) the wire is pretty small gauge, and you should be able to solder that on if you are careful.
Another option might be to use a wire-wrap tool, and wrap some AWG-30 wire wrap wire on each connection - if the pins are long enough to do that; you'd need about 1/4".

I've attached a very brief outline of what my project is about as you weren't sure where I was going to stick the wires. I hpe this makes it clearer. Any suggestions re a 'smarter way of doling it' keeping in mind my very amateur status would be appreciated. My work tend to look uglythough I am trying to improve my soldering skills.
Nobody was born with the ability to solder. ;) It's most definitely something that has to be learned and practiced to be good at it.

My first venture into soldering produced what looked like miniature slag heaps. :rolleyes: It was not pretty.

Here is a link to a short tutorial on soldering:
http://www.elexp.com/t_solder.htm
Note that 63/37 solder not only melts at the lowest temperature, it is also "eutectic"; which means that it goes directly from a liquid to a solid without the "plastic" state. This helps to ensure good solder joints.
 

Thread Starter

faceter

Joined Dec 4, 2011
7
Thanks again SgtWookie for your response.
I'm not sure about the drilling and wire wrapping though:rolleyes:

I have replaced my solder with 67/37 26swg. That should help me improve my work and I will definately being viewing the tutorial.

I may try to move from the Veroboard to and Arduino Prototyping Shield, yet more things to learn:eek:

When finished I may post the completed project if that's allowed to let you cast a professional eye over it.

A couple of final bits, (plse)
The Uno gets 5v via the USB, using the resisters calculated there is 3.5v.
I am adding 4 leds to the circuit (Red, Green, Yellow and Blue) that match the circuit wiring. These are to show (in the dark) that the scope is being moved. The correct resister is in the circuit before the Leds (68Ω and 56Ω) inkeeping with their respective Datasheets.
Do you think there will be enough energy to simulate a button press. I don't have any details relating to the circuits emplyed inside the mount which feed the handset.
Sorry it keeps going on but I am learning so much with your help.
Thanks
 

SgtWookie

Joined Jul 17, 2007
22,230
Thanks again SgtWookie for your response.
I'm not sure about the drilling and wire wrapping though:rolleyes:
Don't let it scare you. I don't know how it is in the UK, but here in the States, we can pick up a manually operated wire wrap tool at a local electronics shop (Radio Shack) for a few dollars; it's actually about the cost of two wire wrap wire spools.

I have replaced my solder with 67/37 26swg.
I think you meant 63/37?
That should help me improve my work and I will definately being viewing the tutorial.
Don't forget, cleanliness is a big part of good soldering. I use isopropyl alcohol, 90% or better, and a small nylon bristle brush to "scrub" the board & parts with. The alcohol lifts dirt and oils away from the board.

I may try to move from the Veroboard to and Arduino Prototyping Shield, yet more things to learn:eek:
That's half the fun of it. :)

When finished I may post the completed project if that's allowed to let you cast a professional eye over it.
Sure.

A couple of final bits, (plse)
The Uno gets 5v via the USB, using the resistors calculated there is 3.5v.
If you are powering the USB via a desktop or tower computer, you can get up to 500mA from the USB port. If you are powering it from a laptop, you may only get 200mA or 300mA from it. In either event, use the power sparingly.

[/QUOTE]I am adding 4 leds to the circuit (Red, Green, Yellow and Blue) that match the circuit wiring. These are to show (in the dark) that the scope is being moved. The correct resister is in the circuit before the Leds (68Ω and 56Ω) inkeeping with their respective Datasheets.[/QUOTE]
You have to be careful here. the IR emitters in the optocoupler need 20mA to turn on the output transistors; I don't know what you are using to supply the optocoupler IR current, but 20mA current is about the limit of many microcontrollers.

The current transfer ratio is pretty low for these optos; for every 3mA input, you only get 1mA out. You're putting in 20mA, so you'll get ~6-2/3mA out. I don't know what it will take to signal your motor on the output side of the optocoupler.

Do you think there will be enough energy to simulate a button press. I don't have any details relating to the circuits emplyed inside the mount which feed the handset.
I have no idea. I would make plans to add the LEDs with current limiting resistors to draw no more than a couple of mA current each, but not install them until after it was working without them. Then if it doesn't work after adding them, you know what the problem is.

In any event, if you wish to make changes, you should draw up a schematic with what you have in mind.

Make sure to give part numbers and other relevant information, like Vf @ current or voltages, etc.
 
Top