Help with Water Pump

[()blivion]

What does uC stand for?

Sorry about the confusion Joe.

As KISS pointed out I was going for "microcontroller". Which is nothing more than a primitive computer in a single chip. Actually I personally tend to use the acronym MCU, but uC/μC is the most commonly found shorthand on these forms from what I have seen. Another good shorthand is just the word "micro", but you have to be careful with that one because it's used for a lot of other things in electronics too. None of the shorthand above would have been easy to figure out with just a Google search. Which is exactly why one should not get to comfortable with shorthand.

 

[alec_t]

Joe, when you build the revised PDM I suggest you allow space around the Q1 base/emitter area to accommodate two extra resistors in case we need to adjust the current-limiting in future. I'm thinking two of the resistor values you already have could be combined in various ways to give a suitable value for a base-emitter shunt resistor to raise the limit from the present 3A to, say, 5A if necessary.

 

[salty joe]

I already built Mk13 up to the 7.5A diode. There's no room by Q1 but that's OK-I have the whole other half of the board to wire up parts. Once it's tuned in, I'll make clean copies.

The 7.5A diodes finally arrived yesterday and I hope to find time this weekend to fire up the soldering iron.

I know I asked this already and KISS answered but I am not clear. Let me please try to word my question better. I assigned pins 11-12,13 to U1d. Pins 12 and 13 will be grounded. Should pin 11 remain disconnected?

Thanks very much.

 

[ronv]

Yes, you are correct. You ground the unused inputs and leave the output open.

 

[salty joe]

Thanks Ron.
The revised PDM is done and seems to work very well. Thanks Alec. Every pump toggled right along. The 50K trimmer is set at 26K. Pumps toggled at 2 sec run/2 sec off as well as 10 sec run/5 sec off.

The only thing that seemed a little off was the alarm-when the trip test is pushed it usually sounds for about 8 sec. One time it only sounded for a couple sec. A few times it sounded for 15 sec or so.

Are there any measurments you'd like me to take?

 

[alec_t]

The revised PDM is done and seems to work very well

Phew! A cautious cheer, then, and it looks like we won't need to modify the current limit .

The only thing that seemed a little off was the alarm-when the trip test is pushed it usually sounds for about 8 sec. One time it only sounded for a couple sec. A few times it sounded for 15 sec or so.

Seems ok. The alarm duration depends on the toggle 'on/off' timing. When the +Vt control signal drops to 0V (i.e. pump 'off') the alarm stops a fraction of a sec later (because C3 discharges via D9/R13). If you monitor the +Vt voltage during toggling you could confirm that.

Are there any measurments you'd like me to take?

Now you can set the trip point.
Set the toggling for ~ 2 sec on/ 5 sec off then reduce the trimmer setting a bit at a time until the alarm trips when the pump starts up. Increase the setting again until the alarm just fails to trip at start-up. Done.

 

[ronv]

I think that could be normal (trip test). The alarm timer also gets reset with the toggle, so depending on when you pushed the button and for how long the time can change. Try it when the pump first starts for just a second or 3, it should be more like 15 seconds then.

 

[salty joe]

Setting the trip point went well.

I pushed the trip test just as the pump came on and the alarm did sound for a longer time, about 12-13 sec.

Thanks you guys for hanging in there and making this happen. A very special thanks to Alec for the outstanding schematic work. Very impressive and amazing electronics one and all.

If any of you guys are ever in NE Ohio and want to stop by for dinner and a drink or two, let me know.

Now all I need to do is copy this little jewel five times and then see how the water moves. Should be fun.
In addition to the six pumps for water movement, I want to use two more of these pumps that will run nonstop. If I provided 12V and the supply V, could this revised PDM be used with D9, R1, R2 and R13 eliminated to run these pumps nonstop?
Thanks a million you guys-you're the best!

 

[KeepItSimpleStupid]

Not really. This is where having identical modules also pays off, so you can swap.

You sort of have
Vt
Alarm
+12
Ground

Now if this were a connector
Vt--------
Alarm....|
+12-------
+12 IN
Ground

The ... are placeholders. The - and | make up an ersatz jumper.

Note that I added another pin for +12. So to use a module continuously, you would just use a different connector. +12 IN, Alarm and Ground are still valid for a pump that runs all of the time. Now this connector would have a jumper between 12 (that was added) and Vt.

Remember that the pumps are matched to the modules, so if you change one, you have to re-adjust.

Or, you need to find a way to use a removable jumper. I think I sent you one in the first bag of goodies.

 

[KeepItSimpleStupid]

Some, if not all of those components are for protection.

Joe, it might be worthwhile to do:

1) Turn the power supply off briefly, like a second and back on. (Simulate a power glitch)
2) Turn the power supply off for a long time like 30 sec and back on (Simulate a power fail)
3) Physically stall the pump for a few seconds.

Repeat one and two when all pumps are connected.

If, I'm correct, the fuses are thermal, so they don't blow.

Good job Joe!

 

[salty joe]

Thanks KISS. It was all you guys. Great idea to simulate a power outage & glitch.

Are you saying to connect V+ to the unused IC pin 12 for the pump to run nonstop?

 

[()blivion]

If, I'm correct, the fuses are thermal, so they don't blow.

Yup, they should be PTC poly fuses.

 

[KeepItSimpleStupid]

It must be getting late for you?

It's Vt. Vt is a 0 V (low/off) or 12 V (high/run) that determines whether or not the pump will run.

Vt HAS to be connected to a LOGIC HIGH in order for the pump to run. +12 is a logic high. The UNUSED pin on the IC is also a logic high, so either will work.

So, what I was saying earlier is that Vth, +12, Alarm, Logic ground are basically LOGIC power and SIGNALS. If you had a 5 pin connector and duplicated +12 (i.e. tied them together), the extra pin becomes a LOGIC high.

Although, I think you built dual pump drivers, so:

LOW, vt1, HIGH, LOW, vt2, HIGH, Alarm1, Alarm2, +12, GND

Would probably make a good start for a connector arrangement. An accidently reversed connector wouldn't likely do much damage. It might not be a good arrangement either.

But, look at it this way, with the board out, you could power it easily or could connect vt1 or Vt2 to LOW or HIGH. Some pins would be unused, but would allow EASY troubleshooting. or would be used in a constant ON mode.

Dumb example, but nonetheless an example. You could have two dangling harnesses (female connectors), that when installed one would run motor A and the other motor B.

Now, it might make more sense to make the ALARM connectors separate. You could have a connector labeled (+12, GND, Alarm1 and Alarm2) on another connector and drop Alarm1 and Alarm2 from the first one.

Don't forget a connector for Vsense1, Vsense1 Gnd, Vsense2, Vsense2 GND
Basically connects across R2. You may not connect this to anything but a meter. Without a connector, it's just a test point.

And then we were all thinking depluggable screw terminals for power in and motors.

I think we are still up in the air as to whether the voltages need to be reduced to 18 V max.

 

[alec_t]

If I provided 12V and the supply V, could this revised PDM be used with D9, R1, R2 and R13 eliminated to run these pumps nonstop?

It depends whether you will want speed control or not. If you will, then keep R1 and R2 and remove D9/R13. If you won't, then keep R2 and remove R1/D9/R13.
Talking of speed control, when you build the other PDM modules allow for the possibility that R8 may need to be increased in value to ~10k, and a wire connection point ('hook') at the junction of R8/D7 will be needed, as per the attached.
View attachment 68562

Edit: See correction in later post

 

[alec_t]

I think we are still up in the air as to whether the voltages need to be reduced to 18 V max.

If Joe opts for a speed control module then, with the SPD (synchronous pulse delay) system I proposed earlier, by turning down the wick the average motor current (and hence coil heating) using a 20V supply can be controlled to be equivalent to the continuous current reulting from using an 18V supply.
I'll post a schematic of a slightly-revised speed module shortly.

Edit: Here tiz

 

[KeepItSimpleStupid]

But "What if" voltage is to blame? Just don't know.

 

[ronv]

Don't forget that if you just tie Vth high there will be no high current pulse on power on so we could be back to a pump that won't start.

 

[alec_t]

Oops. Just spotted an error in my post #1254. Regardless of whether or not speed control will be added it will be necessary to retain R1.

But "What if" voltage is to blame? Just don't know.

We don't know for sure, but IIRC the OEM controller was driving the pump with 21V at some point in its program, so 20.6V shouldn't hurt providing it's not continuous.

Don't forget that if you just tie +Vt high there will be no high current pulse on power on so we could be back to a pump that won't start.

Good catch, Ron. A work-around seems to be to connect a 4k7 resistor between +12V and C4. That results in the current-limit being temporarily raised to ~5.6A ~200mS after power-up, which should give the necessary nudge to get the pump going.

 

[KeepItSimpleStupid]

Why do I know where to look for worms?

 

[salty joe]

Why do I know where to look for worms?

Because you're quite the troubleshooter. Everyone of you guys are clever, creative and brilliant. I am most definitely fortunate and grateful for all your help.

I plan to build the speed control module after I get the six pumps working to simulate the tide and wave. I should finish the second revised PDM tomorrow.

Let me ask this-could a revised PDM built to specs be easily modified to run a pump nonstop?