Grinch and ssr tester

[Daemon]

Hi all

I am trying to build a grinch and ssr tester using the attached diagram which I received off a friend who is really too busy to help me out at the moment.

I have decided I will not need to use L5.

If I take out the part of the diagram between the power and ground wire can anyone tell me if there is anything else I need to include to make it work?
I don't want to connect a wire between the ground and the power do I?


(I have rebuilt this about 4 times because I kept getting it wrong! One of those weeks) I would really like it to actually work this time.

I guess what I am really wanting to know is if I want a wire from the power to the ground if I leave out the L5 resistor and led or should I leave out the wire as well.


Thanks

[TimW]

Hi Daemon




If I take out the part of the diagram between the power and ground wire can anyone tell me if there is anything else I need to include to make it work?


Do you mean leaving out L5 and its current limiting resistor? Just don't include L5 and the resistor. Everything else stays the same. To get the SSR to work you still need to supply the +5 (power) and ground to it via the cable. But its power-power and ground to ground. Not power to ground.



I don't want to connect a wire between the ground and the power do I?


Not unless its so chilly in your part of the world at the moment that you need things around you that are melting?... :)



I guess what I am really wanting to know is if I want a wire from the power to the ground if I leave out the L5 resistor and led or should I leave out the wire as well.


Please leave out any wire directly between power and ground... your power supply (and your Grinch) will be much happier if you don't include it!

Tim

[Daemon]

Thanks Tim

Thats what I thought when I saw the wire between the ground and power originally including the resistor and led.

Would it still have worked out of interest the way its drawn with the resistor and led between the ground and power wires? Would that have made everything sag and melt or would the current have only been able to flow from positive to negative due to the nature of the led?

Anyway. Cheers, I can get on with it now.
Steve

[Daemon]

Hi again

I just cannot get this tester working! It is frustrating me.
I have rebuilt it so many times I am sick of it and tested it to death so it should be working.
Can you please tell me if this sounds right?

These are the steps I have taken.
I have tested the pc power supply output which is 5.1VDC
I have made a tester tester by wiring an RJ45 socket with the ground going to pins 2,4,6,7 & 8 and connecting the power wire to pin 1 of the plug. In this way I should be able to test my tester by plugging it in and seeing all 4 leds light up.

Inside the tester this is what I have soldered.
1. I have soldered a power wire from pin 2 of one plug to pin to of the other and repeated this with the ground.
2. Partway along each of the power and ground wires I have stripped the wire and soldered a wire to each that goes to my strip board(2 seperate places)
3. Where the power wire enters the strip board I have created a bridge from a piece of wire that goes across one end of my 4 resistors(which are 560ohm and I am assured that will be okay)
4. From the other side of each resistor one wire goes to the anode leg of each led.
5. I have soldered wires from pins 2,4,6,8 of the grinch RJ45 socket which each go to the cathode leg of each led.

So this means the power goes from pin 1 to the board through the resistors into the anodes and then from the cathodes back to channels 2,4,6 & 8 of the grinch side


6. Partway along the ground wire I have soldered a wire that goes to my strip board in another area where the ground and power cannot touch.
6. Where the ground wire enters the strip board I have created a bridge with wire that covers the end of 4 wires , each of which goes to one side of the switches.
7. The other side of the switches have a wire going to channels 2,4,6 & 8 on the ssr RJ45 plug.

The switches and ssr channel pins only ever receive ground and the leds and grinch channel pins only ever receive power.

Does anything look out of place there?
I really don't want to rebuild it an eight time.

Things I have tested.
The leds with a watch battery.
Continuity and/or resistance between the power pins of both plugs and the grounds of both plugs
Also between the power bridge and the annode legs
Also between the ground bridge and the switches
Also between the other side of the switches and the ssr channel pins
Also between the cathode legs and the grinch channel pins.

Sorry for the huge post but I am really hoping someone can tell me whats wrong, it is frustrating.

Hopefully someone can help!
Perhaps it might make for a smaller post if someone could pm me with help. I can supply photos if neccessary.

Thanks
Daemon

[TimW]

Thanks Tim

Thats what I thought when I saw the wire between the ground and power originally including the resistor and led.

Would it still have worked out of interest the way its drawn with the resistor and led between the ground and power wires? Would that have made everything sag and melt or would the current have only been able to flow from positive to negative due to the nature of the led?

Anyway. Cheers, I can get on with it now.
Steve

Yes it would have been fine. The difference is one of resistance - the Led and resistor combo would have only passed a small amount of current from power to ground. They offer a resistance to the current so only a sensible amount can flow through them.

A wire between power and ground has a small resistance... so LOTS more current will be encouraged to flow (-> melting overload etc).

In a way most of the electronic components we use (even lights) pass current between power and ground... good designs ensure there is enough resistance to keep the current under control.

[TimW]

Its going to be something simple!

Other things I'd check

1) the polarity of the leds.... the long leg of the led (Anode) connects to the resistors. The short leg (or the notched side of the led= Cathode) conencts to your socket.

2) Make sure you have the orientation of the RJ45 pins right. Many is the time people put pin one where pin 8 should be etc etc!. When you are looking down at a plug with the cable towards you and the locking tab facing down pin 1 will be on the left and pin 8 on the right. When you look INTO a socket (with the locking tab on the bottom), pin 1 is on the left.

If you get the pin order wrong it won't work.

3) Power - you mentioned connecting 'power' to the tester tester. Did you connect the ground from your supply as well?

What I'd do to test is the following:
with the tester tester unplugged measure resistance between power and ground (make sure its not near 0)

plug in the tester tester and measure the voltage from power to ground on the tester. If its not at +5v you need to figure out in the wiring why this is so - something is wrong. If its -ve you have a pin id problem

Once its at 5v leds should start to light... unless you have the polarity wrong.

One last question - what colour are the leds?

Pics always help sort these things out... post away if you're stuck!

[Daemon]

Its going to be something simple!

1) the polarity of the leds.... the long leg of the led (Anode) connects to the resistors. The short leg (or the notched side of the led= Cathode) conencts to your socket. They are connected correctly.

2) Make sure you have the orientation of the RJ45 pins right. Many is the time people put pin one where pin 8 should be etc etc!. When you are looking down at a plug with the cable towards you and the locking tab facing down pin 1 will be on the left and pin 8 on the right. When you look INTO a socket (with the locking tab on the bottom), pin 1 is on the left.Orientation is correct. I checked, double,triple,quadruple and whatever comes after quadruple checked this point. Even dragged the wife out to the garage for a look

If you get the pin order wrong it won't work.

3) Power - you mentioned connecting 'power' to the tester tester. Did you connect the ground from your supply as well?Yes. The ground is connected to pin 7 of the tester as well as to pins 2,4,6 & 8

What I'd do to test is the following:
with the tester tester unplugged measure resistance between power and ground (make sure its not near 0)I get no resistance reading at all between the ground and the power(Thats the same as not getting a zero right?)

plug in the tester tester and measure the voltage from power to ground on the tester. If its not at +5v you need to figure out in the wiring why this is so - something is wrong. If its -ve you have a pin id problemIt measures at 5.1VDC

Once its at 5v leds should start to light... unless you have the polarity wrong.
I get 2 leds lit up with one cable and 1 with another cable. They are both commercial cables!
One last question - what colour are the leds?They are orange.

Pics always help sort these things out... post away if you're stuck!

I think I will need to post some pictures for you. I will try to get some clear ones and post them in a day or so.
Thanks for your help.
Steve

[WireWrap]

...

Once its at 5v leds should start to light... unless you have the polarity wrong.
I get 2 leds lit up with one cable and 1 with another cable. They are both commercial cables!
...



This should raise a red flag! If the cables are the same, then the jacks to which it connects need to be checked for intermittant contacts. More likely is that the cables are NOT the same, especially if the results are consistant over several tests.

Once you find the differences, you'll be near the solution. :D

.

[Daemon]

This should raise a red flag! If the cables are the same, then the jacks to which it connects need to be checked for intermittant contacts. More likely is that the cables are NOT the same, especially if the results are consistant over several tests.

Once you find the differences, you'll be near the solution. :D

.
Yeah. It did raise a red flag but I was too busy puzzling over the testers inability to work correctly. Would a conitnuity check between pins at each end of the cable be enough to confirm it works as it should or are the pins too close to test properly?

[Wayne J]

hmmm, sounds like one cable is a 'crossover' cable, or just has a bad connection. Do a continuity check on them.

[Daemon]

I have used a couple of rj45 sockets to test continuity from pin to pin on 2 cables to be sure it is not the cables. They are 100% correct now.
I also realised while I was checking things that I wired the switches as a DPDT switch instead of a SPST. So I have now wired the ground and channels wires to the same side so that I get no resistance with the switch one way and resistance with it switched the other way.(I am pretty sure thats correct)

Now I seem to get no leds lit at all using the tested cable.?

I thought it was time for some pictures
The first one is of the bottom of the board. The bridge on the right is the power from pin 1 going to the 4 resistors. From the other side of the resistors I am going to the long leg of the leds and then from the short leg of the led to pins 2,4,6 & 8 of the socket.
And the one on the left is the ground. going to the 4 switches and then from the same side of the switches back to pins 2,4,6 & 8 of the ssr side socket.

The second photo shows the resistors. The power comes from pin 1 across to pin 1 on the other side with a connection in the middle to bring off another wire to attach as shown in the picture to the board. This shows continuity from the socket to the bridge at all 4 resistors.

The 3rd photo shows the ground going to the bridge and that is wired exactly the same as the power. The ground goes from pin 7 across to pin 7 and there is a connection in the middle to bring a wire to the board as shown in the picture. There is continuity from pin 7 to the bridge and also between pin 7 and each of the switches.

The last photo shows my socket at the back with the tab on the bottom. I have numbers where I have soldered the channels and power wires.

I rechecked the power at the socket I am using to test and it is still 5.1VDC

I am at a loss now. Any thing that stands out?

[TimW]

Round tracks numbered 10,15 20 ... that's where the resistors are soldered right?

With the power off can you measure the resistance of each of those resistors from one end to the other? It looks like both ends of the resistor are soldered to the same copper track.... that would effectively short the resistor out. The track on the bottom is like a direct wire, so there will be only a very little resistance (if you can measure any across the resistor). More importantly there will be no current limiting... with the power on you'll be measuring 5v across the the leds when the tester is meant to be on (probably fatal for them if I'm right!).

If you have broken the track under the resistors then this won't be a problem (I just cant see it... which is possible!!)

There are easier ways to lay out the circuit on the protoboard you are using. Terminate the resistors on different tracks or even better... across the natural gaps in the middle of the board. That's what its for....

[Daemon]

Round tracks numbered 10,15 20 ... that's where the resistors are soldered right?Correct. 12 14 17 & 20 but close enough :)

With the power off can you measure the resistance of each of those resistors from one end to the other? It looks like both ends of the resistor are soldered to the same copper track.... Hmmm. How come I couldn't see that? :confused: The resistance I measure across each of them is .3-.4 on my meter. What does that mean?More importantly there will be no current limiting... you'll be measuring 5v across the the leds (probably fatal for them if I'm right!).I will test these again and see if I have killed them. That would certainly explain them not coming on :oops:
If you have boken the track then this won't be a problem (I just cant see it... which is possible!!)
Thanks Tim.

EDIT - I have replaced all my LEDS. I had some put aside for another job. They showed as orange earlier probably because I was destoying them. They are actually green.

I just need confirmation now that I have wired the switches correctly.

There are 2 poles each side plus 2 in the center. I have soldered the ground and channel wire to the same side of the switch. So I get no resistance between one side and the other when the switch is turned one way and resistance the other way. Is that correct?

Thanks for your help Tim. I couldnt see the forest for the trees. I now get 4 lights lit up using my testing socket. :D
If I am using external power from my pc power supply, is it still alright to get the power from the db25 cable?

[TimW]

DPDT stands for 'double pole double throw'
It means the switch can 'throw' a connection in 2 directions from the center (double throw) and that there are acrually 2 swictces in 1 (Double pole... each of the center terminals is not connected to each other).

With a DPDT switch you have 6 terminals.... 2 rows of three. Having the swich in one connection connects each center terminal to the terminals on one side of the row... flicking the switch will swap the connection of the center to the other side of the row:


X---X X
X---X X

or

X X---X
X X---X

Does that make sense?

Btw a SPDT switch only has a single pole (so only 1 row of 3 terminals)

X---X X

or

X X---X


Finally if there is only a single throw there will only be 2 terminals in the row

X---X is on

X X isn't

Sounds like you're almost there!

[Daemon]

Cheers Tim. The DPDT switch is the only type I have soldered before now. Your explanation is what I finally figured out.

Did you see my question regarding the power from the DB25 connection?

I am concerned about putting too much voltage into the Grinch. My Grinch board came to me with a jumper on the 2 little pins before the power connectors at the top of the board which I think is for daisy chaining Grinches together. In the wiki it mentions having the power either coming in on pin 1 or for an external source but dont I now have 2 power inputs?

[TimW]

yep.. you only want one power source... if you are powering the board from the PC (and not the db25) just dont jumper the jumper... Dont have both the jumper and power the board separately.

The problem is not too much voltage to the grinch... I'd be more worried if eh two power supplies are not at exactly the same voltage.... you'll get potentially nasty currents flowing between them.

Safer to only connect 1 supply (at one point)

[scorpia]

something that might make testing a little easier is this.

If you happen to have a spare ssroz board and 5 resistors and led's, you could hook up a simple led channel tester as in the picture.
Note, the led in socket 2 is a pain to fit due to the rj45 socket, make sure it is raised a little so it can be bent out the the way of the rj45 socket.

not the picture if of 1/2 of the ssroz board, the rest of the board doesnt need to be populated.

excuse the quick paint cut and paste job but i figured it might be an easy way to show how the led's are located.

[Daemon]

ust dont jumper the jumper...

Sweet. I thought that was what the jumper meant.
I will remove it then.

Cheers

EDIT. I removed the jumper. The led on the grinch still lights with only the DB25 plugged into it and the power supply off. Is that normal? The LED gets brighter when I switch the power on.

[Daemon]

something that might make testing a little easier is this.



Someone suggested I do this when I started. I wish I had of listened to them!

Thanks.