Proportional control of the pump?

So today I've been trying to brainstorm a way to implement a proportional electronic valve for flow control at the outlet of my march pump (to automate matching flow rates for the brutus 20 style CRDFM method), and it just occurred to me that it may make sense to throttle the pump itself via electronic means (something like a router speed control, but with electronic input rather than a knob)

I am sure you would not get a nice simple linear response, but I wonder if it would be good enough to be useful... For something like matching flow rates, you don't need control of the full range from 0-100% of max output - a significantly smaller portion of that range could be good enough.

I wish I had a router speed control lying around to test the idea. I'd go grab a lamp dimmer if I didn't think it would be likely to go up in smoke when used with a motor...

Anyone ever try anything like this?

Funkenjager - I wouldn't do that unless you are using a DC motor. These motors (i.e. March 809) are meant to run at full voltage and it will violate the warranty.

They can actually run significantly hotter on less voltage. The proper way would be to control the volume on the outflow.

Also, there was another post on this recently.
Oh yeah, and a motorized proportional valve would be great

LooyvilleLarry said:

Funkenjager - I wouldn't do that unless you are using a DC motor. These motors (i.e. March 809) are meant to run at full voltage and it will violate the warranty.

They can actually run significantly hotter on less voltage. The proper way would be to control the volume on the outflow.

Click to expand...

Exactly the kind of practical info I needed, thanks...

Oh well, I'll continue my search for a motorized proportional valve, or a way to build one

well, building one is not all that hard, once the math is done

So, you need to figure out how much force is required to move it (while under pressure )> Then you get a servo that will drive that , then some kind of majic connector (lovejoy) that mounts to the valve, then some jerry riggin to get it all together.

If you want proportional you need a needle valve with a V profile. It's going to have to be fairly large too. Then some kind of stepper motor to position it. At least this is how I have been thinking about skinning this beast.

The most practical option though is probably to ditch the march pump and go with a peristaltic pump with multiple head plumbed in parallel to each other. This will get you the flow rates you may be looking for and speed control of the pumps will give you very precise flow and volume control.

Motorized proportional valves are not hard to come by, just not cheap. I have a few hundred Belimo proportional valves in my buildings, both two-way and three-way, controlling hydronic heating coils. And of course you have to have a controller with a 0-10v or 4-20 ma proportional output.

CodeRage said:

If you want proportional you need a needle valve with a V profile. It's going to have to be fairly large too. Then some kind of stepper motor to position it. At least this is how I have been thinking about skinning this beast.

Click to expand...

So far I had been looking at butterfly valves, which seem to be suitable for flow control, and seem to be commercially available in motorized versions (to the tune of $400+).

Of course it's far from impossible to rig a standard butterfly valve with a motor, slap on a feedback pot and set up servo control. But assembling it all so it's sturdy and preferably enclosed and splash-proof isn't trivial - especially if, like me, you don't have access to a machine shop to make the ideal parts.

It would be easier to use a proportional valve like a Belimo, which can be found on ebay for under $100 on a regular basis. Without flow metering you will have fun trying to set and maintain flow rate.

kladue said:

Without flow metering you will have fun trying to set and maintain flow rate.

Click to expand...

Well with the application I mentioned (matching flow into/out of MLT) the real goal is to maintain MLT liquid level, which I have the capability to measure, so I don't need to measure the actual flow rate.

The best way to do it with the pump is to get the 12 or 24 VDC March pumps and then use a VFD (variable frequency drive). You are talking a few hundred bucks. Easier ways to accomplish the same thing might be to incorperate float switches and have the pump come on when needed. A ball valve could get you close and the float switch could be your safe guard so that you can totally walk away from the system.

The best way to do it with the pump is to get the 12 or 24 VDC March pumps and then use a VFD (variable frequency drive). You are talking a few hundred bucks. Easier ways to accomplish the same thing might be to incorperate float switches and have the pump come on when needed. A ball valve could get you close and the float switch could be your safe guard so that you can totally walk away from the system.

PaulB said:

Easier ways to accomplish the same thing might be to incorperate float switches and have the pump come on when needed. A ball valve could get you close and the float switch could be your safe guard so that you can totally walk away from the system.

Click to expand...

I already have liquid level measurement capability with good accuracy, and also have a float switch I could add as a failsafe backup - and the pump is already under BrewTroller control via SSR. For now, I may implement that just to be safe - with a bit of hysteresis it may only end up cycling the pump a few times.
Doesn't help in the other direction though (MLT level falling too low).

Wouldn't it be cheaper and easier to get something like a Blichman Autosparge? It's not a fancy valve and not electronic but it is only $50 and will do the job when flysparging.

My plan is to use a float switch to control the level of sparge water in the MT. That, in combination with an electrially actuated ball valve and you should be able to get good control of the sparge (at least in theory).

Valve

I am building this up now, and hope to have the system up and running late Spring... I can report back when I know how well it will work.

The valves come from China, but from initial observation seem to work well. TBD!

Bryan

Sawdustguy said:

Wouldn't it be cheaper and easier to get something like a Blichman Autosparge? It's not a fancy valve and not electronic but it is only $50 and will do the job when flysparging.

Click to expand...

My reservations with that:
1) it would effectively reduce my MLT capacity since I can only mount it just so high. Someday I'll have a shiny 15 gal MLT, but that's not in the cards yet, so that would be gravity-limiting for 10g batches.
2) I have not been able to find out if it has enough range of adjustment to handle the occasional 5 gal moderate gravity batch in an MLT sized for 10 gal batches.

On another note, I wonder if the autosparge valve turns easily enough to be driven by a hobby servo (with the float arm removed)...

Funkenjaeger said:

On another note, I wonder if the autosparge valve turns easily enough to be driven by a hobby servo (with the float arm removed)...

Click to expand...

Now that's really funny.....

Funkenjaeger said:

So far I had been looking at butterfly valves, which seem to be suitable for flow control, and seem to be commercially available in motorized versions (to the tune of $400+).

Of course it's far from impossible to rig a standard butterfly valve with a motor, slap on a feedback pot and set up servo control. But assembling it all so it's sturdy and preferably enclosed and splash-proof isn't trivial - especially if, like me, you don't have access to a machine shop to make the ideal parts.

Click to expand...

Butterfly valves tend to be used for large diameter piping or when someone is "retrofitting" the valve into piping where it wasn't previously. I'm not sure if it's the best choice in your case.

Although you can, of course, throttle with almost any kind of valve you want, some valve types are more suited than others. I would choose either a needle valve, as CodeRage mentioned, or a globe valve over a butterfly valve, especially if it's a "quick opening" type butterfly valve. Both will give you better performance as you can probably get them in a "linear" variety.

Linear means that your flow increases proportionally to valve lift. So if you've got the valve half open, your flow will be half what it is if it's wide open and so forth. There will be some small variation as your pump head will change as flow changes, but if we're talking 0.85 gpm vs 0.8gpm...then it's roughly constant pressure across the valve.

If you're going to control with an actuator, this easy predictability might come in handy since you would have a roughly linear flow response to your input current/voltage.

A globe valve or similar could in fact be easier to motorize as it's multi-turn instead of quarter-turn...

But if I can pick up a motorized butterfly zone control valve on eBay for a reasonable price, it's tempting, even if it's not optimal...

Take a look at this thread with the 12V floating control ball valve https://www.homebrewtalk.com/f11/march-pump-control-question-151267/. This is something that could be implemented on a brewtroller with 2 digital outputs and a control loop that pulses the valve open/close connections in response to level. I thought the valve price was not excessive and a power supply, couple relays and some programming should be what is needed to make it go. This would be much easier to build and control than pump speed control for level control because brewtroller boards do not seem to have analog outputs to work with.

Funkenjaeger said:

A globe valve or similar could in fact be easier to motorize as it's multi-turn instead of quarter-turn...

But if I can pick up a motorized butterfly zone control valve on eBay for a reasonable price, it's tempting, even if it's not optimal...

Click to expand...

Don't buy one yet. I was going through my Belimo "junkdrawer" and found a proportional actuator. All my ALC DDC units use on-off "floating" actators, so I have no use for the proportional. Send me an email: fletcherhomebrew at yahoo dot com. I'll see if I have a valve body to go with it...

I am pretty happy with the motorized valve I came across. The one from the link that kladue posted. The only issue with it is that the flow restriction is not very linear. If you are used to regular ball valves though it is fine.

I looked into butterfly valves and I could not find one small enough for a reasonable sized home brew setup. They all seemed to start at and inch.

The one I went with seems to work fine for sparge purposes. I am fairly happy with the results. I have two 10gallon batches through my new system and the valve worked great.

Edit: I have a belimo 24v 1/2 stainless floating control valve and actuator I am not using. If anyone wants it PM.

Fletch said:

Don't buy one yet. I was going through my Belimo "junkdrawer" and found a proportional actuator. All my ALC DDC units use on-off "floating" actators, so I have no use for the proportional. Send me an email: fletcherhomebrew at yahoo dot com. I'll see if I have a valve body to go with it...

Click to expand...

email sent

It should not be that hard to implement floating control on the brewtroller platform, you need to drive valve to full open, then set pulse duration, and pulse count to get flow control. A bit of testing would determine the starting pulse count for a base flow rate, and then it could count up or down to control level in response to sensor reading. Not as exact as proportional but achievable with digital outputs in this open loop control scheme.

Funkenjaeger, let me know how this goes for you. I'm thinking of using the Belimo proportional valves with a labjack, but haven't completely made up my mind yet. I'd be interested to hear how this works out for you with the BrewTroll. Good luck.

I have 5 Belimo proportional valves in my new system, they work well but are slow response, 90 seconds to full open, and need to open about 15-20% to begin flowing.

A router speed control does not under-volt the motor.
It pulses 120 VAC to the motor at various rates to achieve speed control.

I don't think it'll harm the motor.
I have a Milwaukee electric drill with a speed control on it that works that way. I've had it for over 30 years and I've used the hell out of it. It's still running strong.

What you may find by attenuating the march pump is that it fails to adequately pump fluid at slower speeds.

There's a certain amount of back flow in that type of mechanism - and worse it's palstic, so there's more yet still because the engineers had to build into the pump head the ability to accommodate all the thermal expansion in plastic. So unless the pump is at full tilt it may not work very well. Or it may just move substantially less fluid.

kladue said:

I have 5 Belimo proportional valves in my new system, they work well but are slow response, 90 seconds to full open, and need to open about 15-20% to begin flowing.

Click to expand...

Your using the labjack U3 to control them correct? Are you using the LJTick-DAC to get the 10v control signal or am I missing how you generate the control signal?

I don't think it'll harm the motor.
I have a Milwaukee electric drill with a speed control on it that works that way. I've had it for over 30 years and I've used the hell out of it. It's still running strong.

Click to expand...

That's apples and oranges. A drill motor is just like a router motor - they're both brushed "universal" types. The motors in the pumps we use (March, Little Giant, etc) are induction motors - no brushes.

The control system that is in use has a bit more capability than the Lab Jack series modules. It is based on Opto 22 PLC hardware and is controlled by an application written in Java. The valves are controlled by 0-10V output modules on the left hand board in the panel, temperature sensors in the middle, digital on the right side. Here is a pre installation shot of the PLC hardware http://picasaweb.google.com/kevin.ladue/Phase2BrewingPanel#5134053722278984738 here is the touchscreen panel PC that runs the Java application http://picasaweb.google.com/kevin.ladue/Panels#5339967114746958306