Johnson pump deal?

[Turbokeu]

Originally posted by Since87
Very good point. I had only been thinking about the diode in terms of clamping the switching spikes while the pump is running.

I hadn't thought about the pump acting as a generator until it came to a stop.

I agree. The diode should go in. (A 1N4001 is probably the most common diode that would work here.)

A 1N4001 is just great.

CD

[Since87]

Does anyone KNOW whether the life of the brushes and seal will be extended substantially by running the pump at lower RPM?

[satanicoo]

I have some motors of 7.2 volts, that i run at 12v, and they last about a year.
I am using an equal 7.2volt motor at stock voltage, for almoust 3 years, still working.

[Since87]

Originally posted by satanicoo
I have some motors of 7.2 volts, that i run at 12v, and they last about a year.
I am using an equal 7.2volt motor at stock voltage, for almoust 3 years, still working.

The situation with overvolting a motor is quite different than undervolting a motor though. I'm not sure this is too relevant.

I would expect the overheating that occurs with an overvolted motor to reduce the life of brushes, bearings, seals, etc.

[satanicoo]

Undervolting it should make it last longer, i just dont know by how much.

[bigben2k]

Well, seeing that I'm probably going to run two of these pumps, along with an automotive blower (I picked up one from a Mercury Grand Marquis, from a used part shop), and that I'd like to keep my PC power lines pretty clean, for an overclock, I'm going to have to use a seperate PSU.

This is still good info though, thanks Since87.

[Turbokeu]

As I told before, I ordered two of these Johnson pumps at Surplus Center.
They were shipped on June, 6th.
Yesterday I received this mail:

Quote

Chris,
Today the Post Office returned your package to us. We aren't sure why, there was plenty of postage. We are going to reship it, I just wanted to make you aware of the possible delay.
Sincerely,
Vanessa Knuth
Surplus Center, Lincoln, NE

End quote

So my pumps came to Belgium and for an unknown reason were sent back to the states...

CD :shrug:

[bigben2k]

Bummer

At least it'll give you time to get the caps and diode


I still owe Since87 a return shipping fee: what's the damage?

[Since87]

Originally posted by bigben2k

I still owe Since87 a return shipping fee: what's the damage?

Forget about it. Not enough to be worth the trouble.

I'm just hoping I soldered the diode in with the right polarity before I sent it back. That's been haunting me, because I'm not sure. The diode is buried in hot melt glue and heatshrink.

Koslov,

If it immediately blows the fuse when you try to use it, a backwards diode will be the reason. Better check it out.

It won't damage anything, (except the fuse) but the glob with the caps and diode in it will need to be replaced.

Sorry in advance if I screwed up.

[koslov]

Thanks for all your work on this pump.

I haven't run this pump yet, but I'm praying the polarity is correct! I won't have a system to use this pump in until I finish my WB (hopefully within a few weeks).

[nyyanks]

Could someone post a drawing or something if the wiring of the caps and diodes to this pump.I'm not much of an electrical engineer so the simpler the better.

Thanks

[Since87]

I just wrote this in another thread:

Gather the following:

The pump
A 0.1 microfarad film capacitor (AKA polyester capacitor)
A 1 amp diode (any voltage rating over 20V will do)
An inline fuse holder
A 3 amp fuse that fits the fuse holder
A male Molex connector.

1. Solder the following together:

the positive lead of the pump
one capacitor lead
the end of the diode with the stripe
one lead of a fuse holder

2. Solder the following together:

the negative lead of the pump
the other capacitor lead
the end of the diode without a stripe
a ground lead (black) of the Molex connector

3. Solder the +12V lead (yellow) of the molex connector to the other lead of the fuse holder.

4. Insulate everything.

5. Install a 3 Amp fuse in the fuse holder, and it's ready to go.

You probably have several groups of four wires coming out of the PSU and going to Molex connectors. Isolate one of these cables, and don't use any of the connections on it for anything other than the pump. (and fans or other insensitive hardware)

[bigben2k]

I ordered my second one today.

Thanks again for the good work Since87: this thread has now made it all the way to Australia!

[Turbokeu]

F1 is a 20x5mm or 32x6.3mm slow 2.5 or 3A glass fuse.
C2 is a 100nF whatever capacitor for spike suppression.
C1 is a polar capacitor (can be a value from some 10's to some 1000's of uF, minimum 25VDC, just to smoothen big current changes in the pump's current draw).
D1 is the flywheel diode, to prevent voltage feedback into the computer's PSU, can be any 1N400x diode.

CD

[Guzzardo]

I'm about to build my first watercooled pc and this looks like the pump I've been waiting for. I was just wondering how this pump compares to others? Also how can I rethread the pump to use 1/2" tubing?

[Gulp35]

I was thinking of building a 12VDC Powersupply for this pump (I am building an external rig so I have room). Would I need the Capacitor and diode assembly that you are talking about? I was also thinking of running 2 fans off it also.

[Khledar]

It's my understanding that the additional circuitry is used to "clean" the power draw, from various interferences shown on since87's osciliscope pics. If you dont have any sensitive devices running on your power supply, and the power supply its self is able to handle the fluxuations in the power draw I doubt it would be a problem. I'm running the pump sans additional circuitry right now, and I haven't had any power problems - It puts a small constant draw on my 12V line but that's about it - I've isolated it to one of the "molex chains" as suggested so the only thing running on that chain is a couple fans and the pump, the fans are cool with it and so it the pump.

[bigben2k]

Originally posted by Gulp35
I was thinking of building a 12VDC Powersupply for this pump (I am building an external rig so I have room). Would I need the Capacitor and diode assembly that you are talking about? I was also thinking of running 2 fans off it also.

It would be a good idea, at the very least, to add the diode.

The interference can eventually puts some wear on the PSU, so anything you can do to lower it will help it live longer. Not significant though: the PSU already has some caps.

[Turbokeu]

Originally posted by Turbokeu
As I told before, I ordered two of these Johnson pumps at Surplus Center.
They were shipped on June, 6th.
Yesterday I received this mail:

Quote

Chris,
Today the Post Office returned your package to us. We aren't sure why, there was plenty of postage. We are going to reship it, I just wanted to make you aware of the possible delay.
Sincerely,
Vanessa Knuth
Surplus Center, Lincoln, NE

End quote

So my pumps came to Belgium and for an unknown reason were sent back to the states...

CD :shrug:

The mail in the quote was dated on June 26.
Today I received a (very friendly) mail from Surplus Center asking me if I wanted my pumps shipped again or not...

I thought they were already on their way to Belgium...

Sniff.....
CD

[bigben2k]

Boy I hope they sort it out!

I got my second one yesterday

I'm trying to see if I can fit them in my blower shroud, since they're supposed to be mounted "heads down" (i.e. motor on top). I'd thermal epoxy a couple of heatsinks that I'll have "bent" to shape...

The problem is that I've only got 10" in depth to work with, and that's supposed to include a 2" thick core, and the PSU array's power rails. The airtrap will have to go behind the fan. The blower wheel is 6 1/4" in diameter, and I was hoping to have at least a one inch clearance around it.

The best fit I've got so far, is to mount the blower a bit lower, where the intake would actually pass over the crossbar, and I'm not sure if that would even clear the mobo!

[Gulp35]

I have a question.

Let's say that the brushings wear out. Are they easy to get at?
Are they cheap to replace? I know my DREMEL has very accessible brushings, is this pump similar to that?

<OT> BB2k your HC blower project has made me want to buy a motor and impeller from HERE and build a shroud/casing for it in a similar way to how #rotor made his composite HC shroud. </OT>

G35

[bigben2k]

From what flyinglass (?) posted, these are infinitely difficult to take apart, but otherwise the brushes are replaceable, if you can find some.

As for the blower, I can only recommend picking something a bit smaller than what I picked up (tip: the Surplus Center has blowers too!). Also, try to find something with a motor that's inside the hub, as much as possible: the 8" width of the case is easily filled, in fact, I won't be able to center this blower over the core.

Otherwise, this automotive blower runs perfectly on a PSU's 5 volt line

[bigben2k]

Ok. Because I'm building a PSU array, and that I don't have room for a 4th power supply (!), I'm forced to rethink the idea of putting this pump on a seperate PSU.

So I need to have a better idea of how running this pump might affect the overclockability of a system. Opinions?


PSU Array

Note also, that I plan on using TWO of these pumps on the 12v line, and an automotive blower, on the +5 line.

[Since87]

There are only two issues that I see.

1. The peak current draw of each pump is likely to be 2A based on looking at the data I gathered earlier and doing some handwaving. A total current draw of 4A is pretty substantial, especially with newer systems which put much more of the load on the +12V rail. Your paralleled supplies shouldn't have a problem with this.

2. The transients may not be as well filtered by the PSU output caps with your setup. With these high frequency transients, the inductance of the cabling probably has a greater role in the transient's effect than the resistance. Depending on what you have for capacitance at the point where all of your +12V rails come together, more of the transient may go to other components (such as the motherboard) than would be the case with a typical PSU. Additional film and/or ceramic capacitors at the point where the pump cable connects to the +12V bus will make up for this. A ferrite EMI suppression core on the +12V line, between the cap/diode combination mentioned earlier, and the PSU output, would probably do an even better job of isolating the transient. EMI suppression ferrites that are easily available, may not be of much use with 3 to 4 Amps of DC current going through them though. (The ferrite might well be saturated at that level of DC current, which would make it ineffective.) It would require some research to find an appropriate ferrite core.

[bigben2k]

Ok, I see what you mean.

The blower is the one I'm really concerned about: although I'll be running it at only 5 volts, I found a similarly sized unit at the same Surplus Center, and this thing will draw 10.5 amps, at 12 volts:
http://www.surpluscenter.com/item.as...tname=electric

I'm also keeping in mind that there's a < ~200 millisecond delay, between the PSU voltages getting to their respective voltages, and the "PWR_GOOD" line coming on.

I need to figure out the amp draw of that blower, at 5V, so I can spec an EMI filter. Ok, time to go through the Digikey catalog again!