How Does This Current Sensor Close its switch?

Status
Not open for further replies.
OK In other projects i connect the wiper (2) to 1 or 3 This setup is different.
View attachment 132703

Yes. In the image above, the pot is wired as a variable resistor.
The pot used in the peak detector is wired as a potentiometer.

i thought these caps did not have a polarity? Im using ( bi-polar cap 22uF 16V Radial Tantalum Capacitor ) I will add the values though.
View attachment 132704

22uf is quite a large value for a non-polar capacitor. Best to verify by looking closely at the body of the cap. There will be a small + sign near the positive pin. Sometimes the + sign is hard to see. Sometimes its just a dot.
 
Added polarity symbols
changed relay diode style
resized CT on PCB
Rerouted nets
Copper pours connect all grounds
---------------------------------------------
I want to thank every member of this group. Including PCB production each module is $4.55 Most importantly i have an understanding of how a CT works. To me that's priceless.

 
OK In other projects i connect the wiper (2) to 1 or 3 This setup is different.

In your other projects, your pot was wired like a rheostat to change the resistance (change current flow). In the method used in this project, the Potentiomer is used as a potentiometer to divide voltage without changing resistance flowing from pin 1 to 3 (assuming a very small amount of current is flowing out of the wiper (pin 2) - which is the case as the wiper is connected to the gate of your 2N7002.

also, take care to finish your traces on the PCB - there are a few yet to be routed (white circles)


 
Also, you can go to technology tab and change the default with for a signal trace. Your traces are very narrow if they are carrying any significant current. You can use the trace-width calculator on under Help if you know the current expected in each area of your circuit.
 
You can use the trace-width calculator on under Help if you know the current expected in each area of your circuit.
thats a great tip. I di not think of increasing the track width. I just dont know what the voltages will be. i will enlarge them as much as i can
 
I added status leds and changed the design of the cap so it would fit in the small box. Most of the track widths are 20,25. As long as it did not break the design check rules i made the as wide as possible.

 
Last edited:
i was letting the copper pour take care of the SMD ground to avoid a VIA
I do not know your CAD program. Why does Q1 have blue pads? Probably ok just it looks odd and please look at the gerber files and see if there are really pads there. Even the ground connection looks odd. (no thermal relief)
 
I do not know your CAD program. Why does Q1 have blue pads? Probably ok just it looks odd and please look at the gerber files and see if there are really pads there. Even the ground connection looks odd. (no thermal relief)
Something to definitely check out. Possibly because the 2n7002 is on the back side of the board?
 
Hi

Some questions and comments.



I took your image and flipped it horizontally so the bottom view is on top.

1. I've show an area outlined in white. If you move this area up, you can then remove the track shown with and "X", then re-route it as show with a white track. This keeps the track away from the CT.

2. I see no purpose for the mosfet Q1 to be a SMD device. Why not just use the standard TO-92 type and not limit yourself to one vendor.

3. If I compare the relay coil markings with the schematic, the pins look swapped. See below.


The datasheet schematic shown is a "Bottom view". So may need to double check the footprint and correct. You''ll need to correct the diode (D1) connections as well.

Comment:

I don't understand why components are mounted on both sides of PCB. Seems they could all be mounted on one side.
 
Last edited:
Reactions: Buk
look at the gerber files and see if there are really pads there. Even the ground connection looks odd. (no thermal relief)
Im using design spark. The SMD (Q1) is according to the software ON THE TOP layer. Pin 2 is the ground. I try to avoid VIAS so if i can (and this board is small) i apply copper pours to top and bottom layers. That connects all the ground pads of each component as a flood. A method a group member showed me. Forgive me If im not using the right lingo.

 
Something to definitely check out. Possibly because the 2n7002 is on the back side of the board?
Im a bit complicated.
The reason is to keep components that have to to with line voltage, adjustments, and connectors are isolated from the circuit. Does this cause any potential problems?
 
1. I've show an area outlined in white. If you move this area up, you can then remove the track shown with and "X", then re-route it as show with a white track. This keeps the track away from the CT.
looks like a last minute play to win by 2 points at the buzzer.
But a great idea too.
2. I see no purpose for the mosfet Q1 to be a SMD device. Why not just use the standard TO-92 type and not limit yourself to one vendor.
If this is the TO-92 version ill get it. the part number and specs are a bit different but they are on the same DS
2N7000 MOSFET N-CHANNEL 60V 0.2A
The datasheet schematic shown is a "Bottom view". So may need to double check the footprint and correct.
Yeah, I thought it was wrong to but the diagram on the DS is bottom a view. I just got a few boards back i designed with the same part and they worked so it is correct.
 
Made all the suggested changes.
That relay drawing is deceiving. if its the bottom view the reference mark should not be there. it makes it appear that your viewing the part from the top. They reference the positive pin as #1 and GND as pin #12. When you apply it to the PCB turning on PIN names and not PIN numbers shows how its orientated. I edited the part in DSpark to read what the pins actually were.
 
The board will be in a box. the connector will protrude out as will the leds. there will be a small hole to access the pot and the top portion of the CT will be visible as well. The rest of the circuit is on the other side to restrict / isolate line voltage from DC.
 
Im a bit complicated.
The reason is to keep components that have to to with line voltage, adjustments, and connectors are isolated from the circuit. Does this cause any potential problems?
Since most of the parts are through-hole, there is no added isolation from placing parts on both sides.

And, assuming that the only line voltage connections is the insulated wire passing through the current transformer, there are no line voltage connection actually on the PCB.
 
looks like a last minute play to win by 2 points at the buzzer.
But a great idea too.

I don't understand your comment.
The track just looked like it could be routed more efficently and kept away from the AC end of the board.

If this is the TO-92 version ill get it. the part number and specs are a bit different but they are on the same DS

Yes...I just didn't see any point to use an SMD device when all the others were through hole devices.
When I searched for the part distributors, OnSem was the only manufacturer offering an SMD version. All others were TO-92.
 
Last edited:

As ChrisP58 mentioned, there are no line voltages on the PCB. In addition, you have the relay mounted at the opposit end of the PCB from the CT....that's good, so no chance the magnetic fields will interact with each other. You can mount all components on the same side of the board if you want to.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…