Help with Water Pump

[salty joe]

double post. oops.

 

[alec_t]

I hope I've got the logic/operation sequence right
A few points to note.
The 6hr and 84 (or 42) sec timing in the first circuit both depend for their accuracy on the clock oscillator frequency. That in turn depends on C8, which is an electrolytic cap and hence has a wide tolerance and can drift in value. I'm assuming the timing isn't critical (after all, there are tidal variations); if it were then a crystal-controlled clock oscillator could be used instead.
I haven't put IC pin numbers on the shematic. You can get them from datasheets. Also not shown are fuses for the pumps.
The CD40106, CD4020 and CD4093 are all static-sensitive. The 4020 is a 16-pin IC.
R3 may need to be adjusted to give the right 'flick' duration. I guess this wouldn't need doing often, so just swap R3 for one of a different value (higher value = longer flick), or else use a trim-pot.

 

[alec_t]

@Ron
Nice timer units. I see from their spec the contacts are rated 10A AC for resistive loads but no max value is given for DC inductive loads, so I'm not sure if they could directly handle a pair of pumps. The minimum load current is 100mA, which is a lot more than the control circuit requires.

 

[salty joe]

Thanks Alec, both the 6 hour and flick times are ballpark. Your schematic looks super.

Is this what's needed for D3, D4, D5 & D6?
Specifications
Product Category: Schottky (Diodes & Rectifiers)
Reverse Voltage: 40 V
Max Surge Current: 80 A
Product: Schottky Diodes

Would a soft start be installed between R4 and U3c as well as between R5 and U3d? If so, would the soft start from post 176 work or would the soft start from post 173 be a better choice?

Should I connect an interference suppression cap between the two terminals of each pump? 0.1μF was used on the first build.

I think I have this right, but need to be sure. C4 gets connected to 12v and ground as well as getting connected across 4020 pins 8 & 16. Is that right?

Could you please give me the critical specs for IRFH3702 so I can maybe find a suitable replacement?

 

[alec_t]

D3-D6 can be the type of Schottky diode you've already got; likewise the MOSFETs. That diode spec above is fine. For the MOSFET, as before you're looking for one with a 40V (at least) rating and a low 'on' resistance (Rdson)(hopefully only a few mΩ).
I must admit I forgot all about the soft start feature when drawing the new schematics . Which one (if any) did you use in your just-built circuit? If the MOSFETS run cool, or only warm, in that circuit then use the same arrangement in the new circuits. Can you cobble the schematics together ok? Let me know if you'd like the combination re-drawn.

Should I connect an interference suppression cap between the two terminals of each pump? 0.1μF was used on the first build.

Yes. Well caught. I should have put that in.

C4 gets connected to 12v and ground as well as getting connected across 4020 pins 8 & 16. Is that right?

Yes, because pins 8 and 16 are respectively the ground pin and +V pin of the 4020. The cap should be physically close to the IC.

 

[salty joe]

IRFH3702 has Rdson 0.0071 ohm (7.1m?) and the 3205 has Rdson 8m ohm.The 3205 is more readily available, but I can get the 3702. Is it worthwhile to chase down the 3702?

No need to redraw anything. I'll just put a note on the schematic. The correct placement of the soft start is between R4 and U3c and between R5 and U3d, is that right?

C4 needs to be close to 4020, thanks. Is it safe to assume that C5 needs to be close to 40106 and C6 needs to be close to 4093? I will definetly build it like that, but I'm curious what happens if they are not close together?

So the ground side of C4 will go to pin 8 of 4020. I can see the next thing I need to do is learn how to read datasheets for these ICs.

 

[alec_t]

IRFH3702 has Rdson 0.0071 ohm (7.1m?) and the 3205 has Rdson 8m ohm

Either would be fine if the voltage requirement is met (haven't checked the datasheets).

The correct placement of the soft start is between R4 and U3c and between R5 and U3d, is that right?

Spot on!

C4 needs to be close to 4020, thanks. Is it safe to assume that C5 needs to be close to 40106 and C6 needs to be close to 4093?

Spot on again.

I'm curious what happens if they are not close together?

ICs, particularly CMOS ones like these, are prone to malfunction if they have electrical noise on their supply. Even a few cm of wire can act as an antenna to pick up interference at high frequencies; hence the rule is keep wires as short as possible and provide these 'decoupling' caps which effectively short out high frequency noise.

I can see the next thing I need to do is learn how to read datasheets for these ICs.

You've learned a lot already, but datasheets are essential reference sources when working with unfamiliar components. In general the most important things you need to check in a datasheet are the component's pin-out and the maximum and minimum values for supply voltage and current. These are usually set out near the start of the datasheet. Looks like you're getting hooked on the hobby

 

[salty joe]

I am kinda getting hooked. Thanks for making it possible.

 

[KeepItSimpleStupid]

CMOS IC's have very good noise immunity because the levels are VCC and GND, but there are TTL compatable CMOS IC's too.

 

[alec_t]

CMOS IC's have very good noise immunity because the levels are VCC and GND

Very true, but I like a belt-and-braces approach so slap decoupling caps on anyway.

 

[KeepItSimpleStupid]

Decoupling caps are ALWAYS a good idea. For those unaccustomed to CMOS, all inputs MUST be tied to VCC or ground. Generally the one that's picked would be the one that would generate the least power dissipation. If you don't, the IC might get awfully hot when it decides to start oscillating on it's own.

 

[salty joe]

Does the little circle on the IC symbols for U2 & U3 represent output?

I don't understand what's going on at the top of U1. Pin 11 on IC4020 is reset, is that what R means?I can't figure out what CP means and don't get what the symbol to the left of CP means.
Pin 10 on IC4020 is shown as a circle with a line through it. Is it a blank pin?
I can't open the file from post 237. Am I missing critical info?
I'm hitting the rack-I'll get back at it tomorrow after work .

 

[KeepItSimpleStupid]

The circles are used to denote inversion. The symbols are typical logic symbols. Nand gates are subject to De-morgan's rules which can mess up the meaning of gates.

R means reset
CP means clock pulse or Phi in the datasheet.

It looks as if the intent is to show the AND gate internal to the IC. Datasheet here: https://www.futurlec.com/4000Series/CD4020.shtml

But it looks like it isn't done quite right. There should be an inversion on pin 10, but not pin 11. My take.

 

[alec_t]

But it looks like it isn't done quite right. There should be an inversion on pin 10, but not pin 11. My take

The standard symbol for the simulation model of the 4020 is confusing. The reset pin (R, pin 11) is active high (I initially mis-read the symbol as meaning active low), so cap C7 pulls it high at power-on. Thereafter R1 pulls it low, enabling the clock pulses at pin 10 to do their thing. So, to clarify the connections required: pin 8 = ground, pin 10 = clock pulse input from the 100k pot / R2, pin 11 = reset pulse from C7/R1, pin 5 = Q5 out at 84 sec (or use pin 7 = Q4 out at 42 sec), pin 3 = Q14 out at 6 hrs, pin 16 = +12V.

I can't open the file from post 237. Am I missing critical info?

Do you mean you can't download it or can't get LTSpice to open it?

 

[salty joe]

Thanks for that hookup plan. I was trying to figure that out and getting nowhere fast.

Do you mean you can't download it or can't get LTSpice to open it? [/QUOTE]

Well, I can get it to open but it is nothing that makes sense to me. The word "wire" followed by a series of numbers. Repeated a whole bunch of times. Different numbers maybe but a column of the word "wire" about 20-30 long.I saw the word "symbol" followed by a series of numbers as well. No option to download when I right click. I'm dying to know what's in that file.

Is it safe to assume that the three +12V "flags" go to the Vdd pins on the three ICs? Is it also safe to assume that it does not matter which "flag" goes to any given IC?

Will the 40106 get hooked up like the last circuit? Moving from the flat side of the triangle to the circle at the point of the triangle 9-8, 1-2, 3-4. Bring 12V to pin 14 then ground all unused odd numbered pins?

BTW, I ran four pumps all day yesterday with the first circuit and not a single component got even warm.

 

[Reloadron]

@Ron
Nice timer units. I see from their spec the contacts are rated 10A AC for resistive loads but no max value is given for DC inductive loads, so I'm not sure if they could directly handle a pair of pumps. The minimum load current is 100mA, which is a lot more than the control circuit requires.

Hi Alec

Yes, just thought I would toss it out there if Joe wanted it or could use it. However, what you have will work fine for the long time delays. I always liked that chip for long timing applications. Also, Joe, please forgive me but you are taking to this stuff kike a fish to water. You guys that are really into the salt water aquariums really have some fantastic aquariums. Totally beautiful fish to watch and other sea creatures. Just incredible and someday please pictures?

Ron

 

[salty joe]

Thanks Ron. You are not so far away that if you felt like it you could see it in person. Plus it would be a good excuse to throw something on the barbeque! Pictures for sure, it will be a little while-late spring early summer. I hope. I've been working on this for maybe two years or so, everything made from scratch including aquarium and all holding tanks. Getting close.

 

[Reloadron]

May take you up on that!

Ron

 

[salty joe]

May take you up on that!

Ron

That would be cool.

I'm having a little bit of trouble finding a .33 uf cap for C10. The couple I found were only available in quantity. Any chance for a more common substitution? Or, I can keep digging.

I know it needs to be at leat 25V, is there an upper limit for V? In other words, would .33 uf 1000V work as well as .33 uf 25V?

 

[salty joe]

Very true, but I like a belt-and-braces approach so slap decoupling caps on anyway.

I like the heavy duty approach. I like it a lot!