DC collector

Thread Starter

seeker28

Joined Mar 29, 2006
8
need help in making a schematic to take DC resistance from multiple traces.
Problem:
Need to take DC Resistance reading from a cable that has 4 rows containing 19 traces each. I wanted to use 19 toggle switches which controls which pins i will take DC from and a 4 position rotary switch which will control what row. Can this be done or the use of more switches has to be used.
 

Papabravo

Joined Feb 24, 2006
21,159
Originally posted by seeker28@Mar 29 2006, 11:46 PM
need help in making a schematic to take DC resistance from multiple traces.
Problem:
Need to take DC Resistance reading from a cable that has 4 rows containing 19 traces each. I wanted to use 19 toggle switches which controls which pins i will take DC from and a 4 position rotary switch which will control what row. Can this be done or the use of more switches has to be used.
[post=15608]Quoted post[/post]​
Your terminology is quaint to say the least. Do I understand that you have a cable with a total of 76 conductors arranged as four bundles of 19 wires each?
 

Thread Starter

seeker28

Joined Mar 29, 2006
8
Yes you are correct, basically I am testing flat cables which are all connected to connectors on both ends. The traces are copper. the method used to take DC resistance was the use of mating connectors with lead wires on both ends. this was time consuming and very disorganized when testing a total of 76 pins. The idea was to build a testing box using a printed circuit board to test the cable using 19 toggles and a rotary switch. Any help is greatly appreciated. Note that i have designed a PCB prototype, but when testing, I was getting the resistance value of all 76 pins. the box would be too big for testing using 76 toggles and since there is 2 sets of male and female connectors, so that i can test different types of connectors...male to male, female to male, female to female...etc.

again thanks for any help.
 

Papabravo

Joined Feb 24, 2006
21,159
Originally posted by seeker28@Mar 30 2006, 01:38 AM
Yes you are correct, basically I am testing flat cables which are all connected to connectors on both ends. The traces are copper. the method used to take DC resistance was the use of mating connectors with lead wires on both ends. this was time consuming and very disorganized when testing a total of 76 pins. The idea was to build a testing box using a printed circuit board to test the cable using 19 toggles and a rotary switch. Any help is greatly appreciated. Note that i have designed a PCB prototype, but when testing, I was getting the resistance value of all 76 pins. the box would be too big for testing using 76 toggles and since there is 2 sets of male and female connectors, so that i can test different types of connectors...male to male, female to male, female to female...etc.

again thats for any help.
[post=15615]Quoted post[/post]​
I'm going to put the rotary switch on the left side. Each of the four output contacts goes to all 19 wires of the corresponding bundle. One probe of the ohmeter goes to the rotary switch common.

On the right hand side connect wire 1 from all four bundles to one side of switch 1. Connect wire 2 from all four bundles to one side of switch 2. Continue in the same fashion until you have wire 19 from all four bundles going to switch 19.
Connect the other side of all 19 switches to a common point. the other probe from the ohmeter goes to this common point.

The test procedure:
1. Let N = 1
2. Switch N Closed, all others Open
3. Cycle rotary switch through all four positions
4. N = N + 1
5. Goto Step 2.

Will this do the job?
 

Thread Starter

seeker28

Joined Mar 29, 2006
8
First I would like to thank you for your quick responses. I have over looked what you said and understand it to the best of my knowledge, I beleive that these pictures might be what you are talking about:







I hope that these pictures help, because the problem that I was getting is that I was getting resistance feedback from all other pins because they were all shorted together. Ideally i would like to get a resistance reading from each pin at a time using the criteria stated previously.


thanks again. your help is greatly appreciated. if there is any more images you may need let me know.
 

Papabravo

Joined Feb 24, 2006
21,159
Insead of "feedback" I think you mean that there are multiple current paths which are upsetting the measurements. I've stared at the drawing for a few minutes, it matches the picture I had in my head, but I don't see where the parallel path is if the switches are wired and set correctly. Do all possible settings behave the same?
 

Thread Starter

seeker28

Joined Mar 29, 2006
8
sorry for the bad terminology, but I am glad you understood what i mean. Here is another image of the schematic showing in red what is happening when taking readings from the cables.



hope this helps.

as you can see the flow of current in the red color.
 

Papabravo

Joined Feb 24, 2006
21,159
Originally posted by seeker28@Mar 31 2006, 06:46 PM
sorry for the bad terminology, but I am glad you understood what i mean. Here is another image of the schematic showing in red what is happening when taking readings from the cables.



hope this helps.

as you can see the flow of current in the red color.
[post=15663]Quoted post[/post]​
Yes, I can certainly see it now. Not sure if there is a fix for this one though.
 

Thread Starter

seeker28

Joined Mar 29, 2006
8
Thank you PapaBravo for all your knowledge, if it cannot be done using 19 switches, how can it without using 76 toggle switches?
 

JoeJester

Joined Apr 26, 2005
4,390
Seeker,

As drawn, you should be able to test each set of cables individually. You'll either read the R value [Cable Open] or 0 ohms [Cable OK].

What is the value of R in your diagram?

Do you read anything with no cables connected and the 19 switches off? Cycle through the rotary switch to confirm your reading infinate resistance. If you have R or R/N, you have either a switch or switches shorted [or reversed in position]. You should be able to cycle through the rotary switch and the individual switches to confirm reading R in each case.

With a known good cable cycle through each connector pair with the same cable. With all the switches open, you should read infinite resistance.

I've seen people reverse a switch accidentally before.
 

Thread Starter

seeker28

Joined Mar 29, 2006
8
Thanks for the response, the first initial test there was a short, i had plugged in the leads from my hp34401A DMM and I was getting a reading with no cables attached, so I checked all my connections to make sure nothing else is shorting out. Now i check it again and now i am getting no reading without any cables, which is good. now i plugged in the cables to be tested and cycle the rotary switch to row A and flipped toggle 1 and my ohm reading was in the 200's Ω . When tested using mating connectors, the R value should be 1.XXX Ω, I have also did some trouble shooting by unpluging all connector rows except row A and I am getting the 1.xxx Ω reading, now i decided to attached another row while still connected and wouldn't you know it the resistance drops and keeps droping everytime you add another row. I understand the the cable has many different trace widths and will give me different Ω reading, but I know what the value of the first trace width is and I have achieved that only when there is no other rows connected to the PCB board. my goal is to either use what I have or add some more switches to make this work where I can only read pin per pin. I am trying to stay away to 76 toggles though. :)
note: the toggles are SPST on-off...any help is greatly appreciated changing the schematic is also welcome..I can etch PCB to meet schematic. even changing the switch is also welcome i.e SPDT, DPST, DPDT.
 

JoeJester

Joined Apr 26, 2005
4,390
What does the backside of that switch panel look like? Do you see resistors inline with the switches?

Below is a diagram of one line [using switches vice the rotary switch and a switch in the cable to simulate an open cable line] of how your diagram is drawn and then another on how you described the operation.

Bear in mind when you flip more than one switch, you are adding resistors in parallel to your meter, hence the resistance should get lower by R / N where R is the single resistor value and N is the number of switches thrown.

As for testing it without a cable, one connector should be a direct short, via the rotary switch and one of the meter test points. the other connector will read R [probably 220 or 270 ohms according to your discussion] when the appropriate line switch is thrown.

With a cable connected, you get infinate ... or 1.xxx k or M ohms, essentially open when the line switch is off, and R when the switch is on. If you turn on a second line switch, you should get R / 2 or 110 or 135 ohms. You really should find out the value of R.
 

Thread Starter

seeker28

Joined Mar 29, 2006
8
Thank you JoeJester, the answers to your questions..... 1st., I have drawn a picture of the back panel and what it looks like since I don't have the part with me, but here it is and No I did not use resistors inline with switches.(should of I used resistors?):





The PCB board will show you how everything was connected.

I know someone will help me get this to work and there is much appreciation to Papabravo and JoeJester.

When testing without cables connected to the box my readings are 0 ohms,

when i connect the cables to the box and flip toggle 1 I get a low ohm reading. As you may have commented:

With a cable connected, you get infinate ... or 1.xxx k or M ohms,
you mention infinate, actually with the cables connected I am getting a low ohm reading. I am not getting kilo ohms nor am I getting Mega ohms.
 

JoeJester

Joined Apr 26, 2005
4,390
Maybe I'm misunderstanding things ... Is that R you have on your first drawing the cable?

If I am misunderstanding, then you should get an open with the switch off or an open cable, and zero ohms with the switch on.

On the infinite k or M [1.xxx] reading, you forgot the essential part ... when the switch is open [off], and the meter reads 0 ... when the switch is closed [on].

No, you don't need resistors. I must have misunderstood your diagram.

ok ...

If your reading all zero's with the switches open, you have an interface problem.

Label the connectors on the outside of your box 1A 1B 2A 2B etc ... until you get them all labeled.

On the wires on the left side of your pcb picture, the cables ABCD and ABCD, connect them to the connectors you labeled B. On the wires on the right side, the ABCD that goes to the rotary switch, connect them to the connectors you labeled A. That should solve your interface problem.

To verify the cables being connected properly, place one end of your meter probe into the test point that's wired to the rotary switch and the other side to the connector that's wired to the area that's on the right side of your PCB picture [actually more like the middle of your pic, since the absolute right is going to the rotary switch. As you rotate the rotary switch, only 1 position should be zero ohms and the rest infinite. Move to the next connector and repeat.

Next move your meter probe to the connector connected to the left side of your PCB diagram. This will take more time because there are 19 connections. select pin 1, then close the pin 1 switch. you should read infinite when the switch is open and zero when with switch is closed. Before moving to the next pin on the connector, cycle each switch to make sure there's no crossed wires. Then move to the next connector pin and repeat the procedure. You will have to repeat the whole procedure for each connector connected to that area. If you see a cross wired condition, remove the other connectors from the circuit first to see if they are influencing your problem ... while monitoring the problem. Troubleshoot and find your problem.
 
Top