Apparently...

 

Well, live and learn I guess. But how did you fry it? Did you forget the load, so that you shorted it when you turned it on? They are static-sensitive, but you need a lot of current to fry one UNLESS you had its gate held in between on and off, from 1-8 volts or so.

 

Well, live and learn I guess. But how did you fry it? Did you forget the load, so that you shorted it when you turned it on? They are static-sensitive, but you need a lot of current to fry one UNLESS you had its gate held in between on and off, from 1-8 volts or so.

I have no idea... It was working fine. I guess it wasn't the one we were supposed to be using...it was an IRF640... I soldered a 1KΩ resistor on to the gate end and it did what it was supposed to. but with one difference, if I touched it to the positive the light came on and stayed on until I touched it to the neg. terminal. But then the light started to come on all by itself.

I dug around and found a IRF510 and promtly cooked it.... I guess I need new glasses!

I will have to look closer at the tiny print! I know I have the IRF540's, just have to de-solder them...

 

The fellow I have hired to do most of my hydro system is also quite an electronics guy. I told him I was edging into his territory and this was his reply...

The fets will do fine for the loads like the leds. You have to bias them right, dont let the gate terminal get higher than the drain terminal. When working right, it only takes a tinsy bit of current to turn them on, so you can limit the gate current with a resistor like a 100, 000 ohm. I have wrecked my share, but once you get the hang of them, they are real handy. but little relays are way more rugged.

If you want AC control, then get some SSR (solid state relays) same as in Baughmans hydro. I use them to turn on kitchen celiing light using a small wall cube plugged into a elcheapo timer. I have the ssr in parallel across the wall switch.

My gawd... it looks like English...

 

...so you can limit the gate current with a resistor like a 100, 000 ohm.

Since there is almost no current, a large resistor will work for low speed switching, and so will NO resistor. A very low resistance allows faster switching. Some resistance is usually added to damp LC ringing of the gate.

... but little relays are way more rugged.

I'm sure he has some experience to refer to but you'll get a lot longer life out of a proper MOSFET circuit. They sit in your TV and switch hundreds of volts at high frequencies all day long for years.

If you want AC control, then get some SSR (solid state relays)

Yup, a SSR is just like a relay, in the sense that you can use it to have a DC circuit control an AC circuit, but it is not a mechanical device.

 

...dont let the gate terminal get higher than the drain terminal

That comment makes no sense to me. When on, the gate is at 10-15V and the drain is very nearly 0V, same as source. So the gate is above the drain frequently in normal operation. It's true that gate-to-source is usually max at 15V (per the datasheet), and you don't want to exceed that.

 

Well, I have to give kudos to those of you who can wrap your heads around this stuff. I have been playing with it and it's not easy!

I thought I was getting somewhere with the MOSFET's and a couple of diodes, but no...

I read a reply on someone else's thread that "we don't do it for you". Well, that's great, because how would we learn if we didn't have to do it!

However, that's why most of us are here, because we need help putting it together in our minds first, then actually making something that works.

That said, I am beginning to have my doubts about this project. If there is a way to do it, there haven't been any simple answers forthcoming. Sometimes we need to admit that we're in over our heads....

 

Maybe a picture will help. Here's an example of using a comparator to control a MOSFET, which in turn controls a DC load. The LM339 comparator actually has 4 comparators on a single IC, and could control all 4 of your circuits.

Each comparator has two states, high or low, depending on the voltages being compared on its inputs. High sends the output floating, that's why there is a 3.3K pull-up resistor on the output. This pulls the MOSFET gate up to +12 when the comparator is high. When the comparator output goes low, it is pulled nearly to ground, turning off the MOSFET. The small current across the 3.3K resistor is easily sinked into the comparator's output pin.

Now, you may or may not end up using a comparator, but it represents any logic scheme that allows you to take the MOSFET gate either high or low.

So that's the power control part of your project.

The other end is converting all your inputs into "logic" signals and then figuring out how to get the result you want. I believe you said toggling on/off. I think each of your input types could be arranged to produce a +12V or 0V signal, right?

I'd be curious to hear from others how to achieve the toggling effect, so that a +12V appearing on any of 4 inputs will change the state of the load.

 

I'm still unsure of what the OP actually needs?

Wayneh (or Zapstrap) are you saying there are 4 switches, and toggling any switch needs to produce the result of toggling the load on/off?

 

That's my understanding of it, that each of 4 input types can produce a +12V signal into the controller that needs to be defined. The OP said he wants a toggle, so that a change in any of those 4 inputs should change the state of the load. I guess that includes going from +12 to zero, not just from zero to +12.

Personally, I think an OR might be better. (And easier to implement, at least to my thinking.)

 

I'm still unsure of what the OP actually needs?

Wayneh (or Zapstrap) are you saying there are 4 switches, and toggling any switch needs to produce the result of toggling the load on/off?

[[[[That's my understanding of it, that each of 4 input types can produce a +12V signal into the controller that needs to be defined. The OP said he wants a toggle, so that a change in any of those 4 inputs should change the state of the load. I guess that includes going from +12 to zero, not just from zero to +12.

Personally, I think an OR might be better. (And easier to implement, at least to my thinking.) ]]]]

Thanks for keeping my idea alive guys.

Frankly, how it's implemented isn't too important. The simpler the better eh? If an OR works best and I can build it out of my junk box, awesome!

 

Hmm, ok so the idea is to make it so when ANY switch is changed, the output will toggle. That lets you turn the output on/off from any switch.

In a 2 switch system that can be done by 2 SPDT changeover switches, as done in 2 switch hallway lighting.

With more than 2 switches it can be done with a DPDT switch at every node, so each node basically reverses the two wires. The first and last switch can be SPDT types, all the middle switches need to be DPDT types.

This will work fine but means of course the power must go in series through every switch. Since you are talking about high current DC power, I would send a low current control signal through the series switches, then use that signal to control the main power relay on/off.

Anyway that's a simple reliable solution, using a switch or relay for each node, and one large power relay at the end to switch the main load.

 

As before, my replies are in red...

Hmm, ok so the idea is to make it so when ANY switch is changed, the output will toggle. That lets you turn the output on/off from any switch. Yes, that's the goal.

In a 2 switch system that can be done by 2 SPDT changeover switches, as done in 2 switch hallway lighting.

With more than 2 switches it can be done with a DPDT switch at every node, so each node basically reverses the two wires. The first and last switch can be SPDT types, all the middle switches need to be DPDT types. I have played with these 3 and 4 way switches in my AC power.

This will work fine but means of course the power must go in series through every switch. Since you are talking about high current DC power, I would send a low current control signal through the series switches, then use that signal to control the main power relay on/off. This is sounding promising!

Anyway that's a simple reliable solution, using a switch or relay for each node, and one large power relay at the end to switch the main load.

Excellent sir. Thank you.

Here's a short video I just made to help show what I'm working with...

 

Well, Merry Christmas everyone...

And dang, you got to be quick on here because this is one busy forum!

I decided to just wire the thing up, sans MOSFET's, and see what happens. Thankfully, nothing melted or let blue smoke out so that's good. I guess I'll just have to be happy with being able to switch these circuits one at a time, since toggling seems to be beyond my ability at this time. I have employed a relay for each of the manual switches. The 12V remote has common ground leads, so I was able to eliminate all but the main one.

I'm glad this forum is here so noobs like myself can get some sort of clue but I only have a few weeks off to get this done and I'm running out of time.

If wayneh or THE_RB, or anyone else for that matter, has any ideas, please feel free to share. Meanwhile, it's time to start drilling holes and fishing wire...

 

Hi again, if you have relays for each node, you can do the multi-master toggling the way I suggested using two SPDT relays for the first and last nodes, and DPDT relays for the nodes in the middle.

 

Hi again, if you have relays for each node, you can do the multi-master toggling the way I suggested using two SPDT relays for the first and last nodes, and DPDT relays for the nodes in the middle.

...If ignorance is bliss then I'm freeking ecstatic!...

Seriously, thanks for the come back. I read what you wrote and saw what you said, but I haven't got a clue! haha...

As per the "mission statement" of this forum, a dumb question asked...

What is a node?

WTF is multi master toggling?

And google has shown me what the Single Pole Double Throw and Double Pole Double Throw relays are...

My gawd Jim, where do they all come from!

Edited to add: I just pawed through my junk box and there ain't no DPDT relays in thar...

 

After downloading a nice CAD program to draw my circuits on, I discovered it would take me a month to learn how to drive it! So, I got out some coloured pencils, paper and a ruler... old school eh...

It's pretty straight forward, it's a rendering of the circuit I made on the test bench.

If anyone has any ideas on how to improve the design, I'm all ears.

As for the SPDT & DPDT relays, I still haven't wrapped my head around that, but it does sound like the way to go. I just need to figure it out.

 

Ok, it's been a while but we have overcome the obstacle of ignorance, somewhat.

I found a nifty little gizmo made for car alarms. It's called "Your Valet, 611T", made by DEI. Basically, it is made to toggle 12VDC loads from multiple sources. It also has a timer feature for up to 100 seconds.

http://alfsoftware.net/docs/611t_timed_latch_relay.pdf

And here is the (almost) finished 12V panel with the 3 611T's on the lower right, just above the 4 channel remote control...

So, without going into detail because I'm rather pressed for time right now, it toggles on all of my circuits, just like I wanted.

That said, this can now go back down the rabbit hole of this site......