Help with Water Pump

[alec_t]

I'd think if vasiline is reccomended for car battery terminals, it'd be because it does not conduct. Generally speaking, anything oil base or grease is non polar.

I'd think so too. But the idea seems to be to smear the battery post with the stuff and put the clamp over the film of goo. Clearly that doesn't prevent 100A or so flowing. Which is why I query its conductivity (a bit messy to measure). Even if it's an insulator I would be concerned that it might creep between the IC pins and the socket contacts over time, thus preventing good contact.

 

[salty joe]

You're probably right. I put too much work soldering this thing together to make a dumb mistake.

 

[salty joe]

Just wanted to post pictures of this circuit. It's all done except for the conformal coating. I should have planned for the LEDs. I ended up having to put resistors for the LEDs in line with the pump leads instead of mounting the resistors on the board. I got it worked out with shrink tubing-it should be OK. I used ()blivion's method of using component leads and bare copper wire for most connections. I like that method a lot more than the jumbled mess my first try was.

I always heard that DC current drops off in a hurry when a long wire is used. So I was pleasantly surprised when 20 feet of 14 gauge speaker wire did not show a drop in voltage.Which means I can place the power supply outside the fish room.

Using a tin as a container worked out like a champ. Thanks for that good idea Alec. Thanks also for answering all my questions along the way. Very interesting experience.

 

[Reloadron]

Congratulations Joe and many kudos to Alec for a fine circuit. Joe, working from Alec's fine instructions and help you did a really great job.

Ron

 

[salty joe]

Thanks Ron. Like you said

many kudos to Alec for a fine circuit. Alec's fine instructions and help.

Ron

 

[alec_t]

Glad to have been of help. Looking forward to seeing pics of the populated reef tank.

 

[salty joe]

It will be a pleasure to post a picture of the tank. I am working out plumbing bugs right now, their is still an awful lot to be done. Hopefully before summer there will be saltwater and live rock in the tank. Thanks again Alec.

 

[()blivion]

Ah, that looks alright for a first try salty. Nice work.

I'd think if vaseline is recommended for car battery terminals, it'd be because it does not conduct. Generally speaking, anything oil base or grease is non polar.

Vaseline (generically called "petroleum jelly") does NOT conduct a realistic amount of electricity. It is effectively an insulator. It is put on car battery terminals to act as a barrier to oxygen/water and to prevent the lead from corroding. Whilst simultaneously being thin and mobile enough to allow a good metal to metal contact when the cables are connected. It can be used on electrical connections provided there is enough force to cut through the surface tension and make a metal to metal contact.

As for your conformal coating. I would put a socket where your IC is. Then punch the IC's pins through the sticky side of some strong tape, going long wise and leaving some excess on each end. One inch wide "Filament Strapping Tape" would be ideal for this. Then fold the tape over the top of the IC and pinch it down all around it. you may want to cut off some of the excess from the sides. This would give you a tab on one end of the IC you could pull pretty good when you want to remove it. Then you can dip the board in your make shift conformal coating and dry it. With this you should be able to remove the IC in an emergency event. The IC *MAY* not want to seat very well in the socket like this however. So you may need to use quite a bit of force to install it at first. Anyway that's what I would do.

 

[salty joe]

Ah, that looks alright for a first try salty. Nice work. .

Thanks. As I look at it, I do see it's a little sloppy. But I feel confident that there is not a single cold joint. I learned how to solder a pinhead sized mound just in time at the end of the build for the IC socket. I wish I could say that this is my creation, but all I did was follow Alec's outstanding instructions with some solid help from you as well.


.[/QUOTE]
As for your conformal coating. I would put a socket where your IC is. Then punch the IC's pins through the sticky side of some strong tape, going long wise and leaving some excess on each end. One inch wide "Filament Strapping Tape" would be ideal for this. Then fold the tape over the top of the IC and pinch it down all around it. you may want to cut off some of the excess from the sides. This would give you a tab on one end of the IC you could pull pretty good when you want to remove it. Then you can dip the board in your make shift conformal coating and dry it. With this you should be able to remove the IC in an emergency event. The IC *MAY* not want to seat very well in the socket like this however. So you may need to use quite a bit of force to install it at first. Anyway that's what I would do.[/QUOTE]

I like that idea. I just might try it. Thanks.

I wonder if you, Alec and anyone else would be willing to help with two more circuits. Same pumps. I'm looking for something similar to the schematic in post 20, except one pair of pumps runs for about six hours then the other pair kicks on for about six hours. Only one pair of pumps would be running at a time. I'd also like to be able to make the pumps run about 30 seconds and off about 5 seconds during the pumps 6 hour run time. This is to mimic the tide. But the kicker is the idle pumps would need to "flicker" or spin just a turn or three every minute or so. Otherwise a fish is bound to sleep tucked against a pump and when it fires up, goodbye fish.

The other circuit would control two pumps. One pump would run for 1-10 seconds, then the other pump would run for 1-10 seconds. It would be important that both pumps ran for the same amount of time. Only one would run at a time. This is to create a wave and really get the water bouncing up and down.

If this is asking too much, I understand completely. I wish there was something I could do for you guys. My offer to do my best to help with any remodel questions is good for anyone on this forum.

 

[Dean Huster]

If you're going to all the trouble of conformal coating, you need to note the definition of any socket with X number of pins: "X potential intermittents or open connections". Ever see NASA use sockets on anything? Tektronix used to be the socket king of the industry, using them for all ICs and transistors; boards were interconnected with sockets/plugs. A good 50% of the instruments we received in the Service Center whether old equipment or warranty equipment were in for intermittent problems or for "permanant" problems that were "repaired" by moving a component in a socket. Technicians loved sockets for the ease in pulling transistors for a quick check on the curve tracer. Poor reason for having sockets. In 1982 when Tek introduced the 2200-series of scopes, the company went through a major change: no more component sockets, vastly-reduced sockets for interconnects. Result? Vastly-lowered construction costs and for other than mechanical switches, intermittents almost disappeared. Always solder ICs to the board if you want reliability. Another bennie is better heat transfer from a high-dissipation IC.

 

[salty joe]

I was mostly worried about cooking the IC to death.

BTW, am I correct to assume anything with an IC should be plugged into a surge protector? How many joules would a surge protector need for the kind of circuit I just built? It has a CD40106 IC.

 

[alec_t]

The 6 hr period rules out a simple CD40106-based timer, but I'm sure we can come up with something using that IC for the shorter periods timing plus a counter for the 6 hr timing. I'll give it a go if no-one else jumps in first.

Re the surge-protector, it would do no harm to use one. The Joules rating is more a matter of how much incoming spike energy you hope to absorb than what the protected circuit consists of. For this project it's not critical.

@Dean
Good advice. However this project was the OP's very first go at soldering, so use of a socket for the IC was suggested to improve the chance of IC survival

 

[Reloadron]

Joe and Alec

Joe, it's up to you but if you want a small commercial timer for the longer 6 hour application I have several and you are welcome to one. For the bumping possibly Alec could piggy back something on the outputs of what I have laying here. However, if you want to have another building opportunity by all means have at it.

Ron

 

[alec_t]

Generous offer, Ron. If Joe opts for that can you post a link to the timer spec?

 

[Reloadron]

Sure, up to Joe and nearly as I can figure I could walk out in my yard and toss the thing and with any luck hit his yard. Seriously, maybe 50Km from me if that.

Ron

 

[alec_t]

OK. No other takers, so here's my suggestion for the first new circuit requested. Parts of it should look familiar, being similar to your present circuit.
The left third or so is the 24V and 12V power supply part (or you could use what you already have).
The middle third or so is the timing and logic part. U2a forms the clock oscillator with a period of ~2.6 seconds. U2b forms an oscillator to give the 30 sec on / 5 sec off periods. U1 counts the clock pulses and its Q14 output changes state every 6 hrs to define the pump active and inactive periods. Its Q5 output goes low every 84 secs (could be changed to 42 secs) and is used to generate a brief (~ 0.1 sec) pulse for 'flicking' the pumps. U3's gates combine the various signals to provide two driver signals: one for the first pair of pumps and the other for the second pair.
The right third or so is the power FET switching part for the pump pairs.

 

[alec_t]

....and here's the much simpler second new circuit.

 

[Reloadron]

Alec & Joe

This is what I have if it can be used in the scheme of things.

Ron

 

[salty joe]

Joe and Alec

Joe, it's up to you but if you want a small commercial timer for the longer 6 hour application I have several and you are welcome to one. For the bumping possibly Alec could piggy back something on the outputs of what I have laying here. However, if you want to have another building opportunity by all means have at it.
Ron

Thanks Ron-that is really generous. Another build is what I want though. Want to do a little neater and cleaner job. Plus I can hardly wait to use my gadget.

Sure, up to Joe and nearly as I can figure I could walk out in my yard and toss the thing and with any luck hit his yard. Seriously, maybe 50Km from me if that.
Ron

You must have an arm like a rocket!

OK. No other takers, so here's my suggestion for the first new circuit requested. Parts of it should look familiar, being similar to your present circuit.
The left third or so is the 24V and 12V power supply part (or you could use what you already have).
The middle third or so is the timing and logic part. U2a forms the clock oscillator with a period of ~2.6 seconds. U2b forms an oscillator to give the 30 sec on / 5 sec off periods. U1 counts the clock pulses and its Q14 output changes state every 6 hrs to define the pump active and inactive periods. Its Q5 output goes low every 84 secs (could be changed to 42 secs) and is used to generate a brief (~ 0.1 sec) pulse for 'flicking' the pumps. U3's gates combine the various signals to provide two driver signals: one for the first pair of pumps and the other for the second pair.
The right third or so is the power FET switching part for the pump pairs.

Thank you Alec. You must be a genius. For real.

....and here's the much simpler second new circuit.

Thank you again. You are without a doubt a very generous person. I've got the plumbing bugs worked out, I'll get these built and use them to work out the water movement scheme. This thing is gonna rock. Thank you for your help.

I don't know why, but I did not get notified there were responses on this thread. I almost was gonna cry. I got on here for the heck of it and it was like Christmas morning!!! Thanks Alec, Thanks Ron you guys made my day.

 

[Reloadron]

OK cool and best of luck on this one, I know it will come out fine.

Later
Ron