# Designing Expansion PEX Layout and Brands to Use

### Users who are viewing this thread

#### Gsmith22

##### Active Member
Great, thank you for the rule of thumb.

Cheers, Wayne
head loss is a function of the fluid velocity squared which doesn't appear in that (linear) rule of thumb equation. so that equation might be a fine for very small flows to get you an order of magnitude head loss but most rules of thumb are conservative which means it will probably get you a pump that is larger than you need, meaning the velocity of the fluid will be more than is necessary creating more head loss because of the whole velocity squared thing in the head loss equation. it is essentially a self fulfilling prophecy. I recommend estimating fluid flow (or velocity) so head loss is much more accurate when you go to use the pump curve. if darcy weisbach formula is too cumbersome there are plenty of charts with head loss/ft of pipe for different pipe types, diameters, fluid flow/velocity and resulting head loss that you could reference for pump sizing.

#### John Gayewski

head loss is a function of the fluid velocity squared which doesn't appear in that (linear) rule of thumb equation. so that equation might be a fine for very small flows to get you an order of magnitude head loss but most rules of thumb are conservative which means it will probably get you a pump that is larger than you need, meaning the velocity of the fluid will be more than is necessary creating more head loss because of the whole velocity squared thing in the head loss equation. it is essentially a self fulfilling prophecy. I recommend estimating fluid flow (or velocity) so head loss is much more accurate when you go to use the pump curve. if darcy weisbach formula is too cumbersome there are plenty of charts with head loss/ft of pipe for different pipe types, diameters, fluid flow/velocity and resulting head loss that you could reference for pump sizing.
The sizing rule of thumb for hydronic systems is generally pretty accurate. If you've ever used a system syzer wheel (available from a bell and gosset rep) there you can easily get the fluid velocity based on these and other inputs. You pretty much always end up within design parameters. There are some things that end up in the margins where you can end up with something you didn't expect using thr rule of thumb. But if you look the rule of thumb it's a sliding scale. Anywhere from 0.03 with pex only(very few fittings and valves, to 0.06 for iron pipe and copper with more fittings and other restrictions.

If hydronics circulators were routinely oversized the velocity noise in the piping and radiators would be evident and probably unlivable. The industry wouldn't use it and the systems/ components wouldn't routinely last 50 years or more.

Last edited:

#### Wren

##### Member
It looks like about 2.2 GPM if I'm reading the chart correctly.

#### John Gayewski

It looks like about 2.2 GPM if I'm reading the chart correctly.

View attachment 80868

Are you trying to keep full hot water in the line? Or just warm? That will definitely keep the line warm, but it's a little low. However pump take a pretty decent jump after that size. I think you have the right pump. I like the taco pumps, but that's just me.

3 or 4 gpm would be what I shot for. But i think you'll be happy.

#### wwhitney

75' of 3/4" PEX holds only 1.4 gallons. So at 2.2 gpm, the transit time for the water from the tank will only be 40 seconds round trip, or presumably 20 second to the farthest fixture. I wouldn't think the hot water would cool off that much in 20 seconds, particularly as the lines should be insulated.

Cheers, Wayne

#### John Gayewski

75' of 3/4" PEX holds only 1.4 gallons. So at 2.2 gpm, the transit time for the water from the tank will only be 40 seconds round trip, or presumably 20 second to the farthest fixture. I wouldn't think the hot water would cool off that much in 20 seconds, particularly as the lines should be insulated.

Cheers, Wayne
You could be right about that. Insulated pipe could be a huge factor in this. Generally 3 to 4 gpm in an uninsulated system would get you about 15 or 20 deg f temp drop. Depending on other factors.

I would be interested for the poster to report back on this.

He's about 1.7 fps also if anyone's interested. So more would be fine, and probably a little better.

Last edited:

#### Wren

##### Member
Are you trying to keep full hot water in the line? Or just warm? That will definitely keep the line warm, but it's a little low. However pump take a pretty decent jump after that size. I think you have the right pump. I like the taco pumps, but that's just me.

3 or 4 gpm would be what I shot for. But i think you'll be happy.
I'm not looking at keeping the water hot all of the time. I'm interested in having the pump come on when the light switch in the bathrooms is switched on. Lights go off and the pump goes off. The full bath/laundry room only has one window and the light would have to be turned on whenever someone is in there as that part of the house is very shady.

I'm planning on insulating the hot/return lines.

If you think that pump will be a good choice, I'll go with it.

#### John Gayewski

I'm not looking at keeping the water hot all of the time. I'm interested in having the pump come on when the light switch in the bathrooms is switched on. Lights go off and the pump goes off. The full bath/laundry room only has one window and the light would have to be turned on whenever someone is in there as that part of the house is very shady.

I'm planning on insulating the hot/return lines.

If you think that pump will be a good choice, I'll go with it.
I would find one that pumped just a little faster, but the selection of pumps looked like it jumped quite a bit after yours. If my math is right (which my math is sometimes shoddy) to get from the heater and back would take about 45 sec. So 22.5 sec to get to the farthest sink.

Finding the perfect pump might not be worth it to you or maybe it is. You have all of the info you need to decide for yourself. I think Wayne thinks your pump will be fine and it probably will. This is more a matter of opinion rather than right or wrong.

#### wwhitney

It looks like about 2.2 GPM if I'm reading the chart correctly.
So this is where, having selected a pump via the rule of thumb (or maybe randomly : - ) it could be worth doing a more exact calculation. If we assume that with fittings the equivalent length of the PEX is 100 ft, and the PEX is 3/4", it has an ID of 0.681". Per http://irrigation.wsu.edu/Content/Calculators/General/Pipeline-Pressure-Loss.php then the pressure loss will only be 2.7 ft of head, not 4.5 ft.

That means the flow will be more than 2.2 gpm. You can try higher flow rates, like 2.6 gpm or 2.8 gpm, until you find the one where the pressure loss calculation is close to the pump lift per the pump curve. I'm guessing 2.8 gpm is about right.

Cheers, Wayne

#### John Gayewski

So this is where, having selected a pump via the rule of thumb (or maybe randomly : - ) it could be worth doing a more exact calculation. If we assume that with fittings the equivalent length of the PEX is 100 ft, and the PEX is 3/4", it has an ID of 0.681". Per http://irrigation.wsu.edu/Content/Calculators/General/Pipeline-Pressure-Loss.php then the pressure loss will only be 2.7 ft of head, not 4.5 ft.

That means the flow will be more than 2.2 gpm. You can try higher flow rates, like 2.6 gpm or 2.8 gpm, until you find the one where the pressure loss calculation is close to the pump lift per the pump curve. I'm guessing 2.8 gpm is about right.

Cheers, Wayne
If someone were confident enough to use the Pex rule of thumb 0.03 then you'd end up with 3 ft head. And you had 2.7.

#### John Gayewski

So this is where, having selected a pump via the rule of thumb (or maybe randomly : - ) it could be worth doing a more exact calculation. If we assume that with fittings the equivalent length of the PEX is 100 ft, and the PEX is 3/4", it has an ID of 0.681". Per http://irrigation.wsu.edu/Content/Calculators/General/Pipeline-Pressure-Loss.php then the pressure loss will only be 2.7 ft of head, not 4.5 ft.

That means the flow will be more than 2.2 gpm. You can try higher flow rates, like 2.6 gpm or 2.8 gpm, until you find the one where the pressure loss calculation is close to the pump lift per the pump curve. I'm guessing 2.8 gpm is about right.

Cheers, Wayne
Did your happen to look at the velocity with these inputs? Still 1.7fps? I'm just curious what you think of that speed?

#### wwhitney

Did your happen to look at the velocity with these inputs? Still 1.7fps? I'm just curious what you think of that speed?
No. Is there something to check or some import, beyond ensuring it's below the maximum recommendation for the pipe material?

Cheers, Wayne

#### John Gayewski

No. Is there something to check or some import, beyond ensuring it's below the maximum recommendation for the pipe material?

Cheers, Wayne
Well if I'm right (I'm only 90 percent sure) then thinking about this guys system. His pump comes on with the light switch. It'll take 20 sec (roughly) for hot water to get there. I think that's a little weak. Just wondering what your thoughts were on this.

#### wwhitney

Just wondering what your thoughts were on this.
Ah, I've never plumbed a recirculation loop, and my energy code would require the whole loop to be insulated and the pump only activated by a separate momentary switch, so I don't have any performance expectations.

20 seconds doesn't sound unreasonable, maybe it's at the edge of acceptability. But I think the calculation that gave us 20 seconds may have been a slight overestimate.

If the pump comes on with the light switch, does it stay on the whole time the light is on? That seems excessive, I would think there should be a mechanism for the pump to shut off when the water temperature at the fixture is adequate.

Personally, with the short distances involved, I don't see recirculation as worth it. If each of the runs from the water heater to a bathroom is 25' and done in 1/2" PEX, that's a volume of under 1/4 gallon that you'd have to wait to drain to get hot water. Of course, you'd be wasting 1/4 gallon of water each time.

Cheers, Wayne

#### Wren

##### Member
If the pump comes on with the light switch, does it stay on the whole time the light is on? That seems excessive, I would think there should be a mechanism for the pump to shut off when the water temperature at the fixture is adequate.
My thoughts about having the pump run whenever the bathroom light are that it wouldn't be an excessive amount of time as some systems run the pumps for hours a day when using an adaptive, pre-programmed controller.
I would be happy using some sort of relay system whereby the lights coming on would trigger a relay that would come of for long enough to get the hot water to the bathrooms. I don't have any experience with that sort of setup as used on 110V systems, although I've used relays in automotive applications. Perhaps using some sort of thermostat tied into the light switch and/or relay would be an option?

Also, if the pump that we've been discussing isn't going to be adequate, I can return it for something with more capacity, I would just need to know what to order.

Thanks for the continuing discussion!

Last edited:

#### John Gayewski

My thoughts about having the pump run whenever the bathroom light are that it wouldn't be an excessive amount of time as some systems run the pumps for hours a day when using an adaptive, pre-programmed controller.
I would be happy using some sort of relay system whereby the lights coming on would trigger a relay that would come of for long enough to get the hot water to the bathrooms. I don't have any experience with that sort of setup as used on 110V systems, although I've used relays in automotive applications. Perhaps using some sort of thermostat tied into the light switch and/or relay would be an option?

Also, if the pump that we've been discussing isn't going to be adequate, I can return it for something with more capacity, I would just need to know what to order.

Thanks for the continuing discussion!
I think the pump you have will be fine. It's just a little slow. Bit that's just my opinion. Controlling the pump would be something you should decide on. Its what works for you and keeps you content. I don't even think you need a relay. The pump should be 110.

#### wwhitney

Also, if the pump that we've been discussing isn't going to be adequate, I can return it for something with more capacity, I would just need to know what to order.
In your last diagram the loop is pretty short, 25'?. It doesn't include the half bath, and just gets water to the main bath quicker (which would also cut the wait time to the laundry). I'd be inclined to just put in 1/2" lines and see what the wait time is. Then if it is a problem, install a preengineered solution like the Chili Pepper pump under the main bath lav. You could install a dedicated 1/2" return line for future use with the under sink pump if you dislike the idea of pumping water back into the cold water supply.

For a tangentially related thread that discusses in passing a couple of the under sink pumps, see https://terrylove.com/forums/index....-starts-to-run-when-cold-tap-turned-on.97224/

Cheers, Wayne

#### Jeff H Young

My thoughts about having the pump run whenever the bathroom light are that it wouldn't be an excessive amount of time as some systems run the pumps for hours a day when using an adaptive, pre-programmed controller.
I would be happy using some sort of relay system whereby the lights coming on would trigger a relay that would come of for long enough to get the hot water to the bathrooms. I don't have any experience with that sort of setup as used on 110V systems, although I've used relays in automotive applications. Perhaps using some sort of thermostat tied into the light switch and/or relay would be an option?

Also, if the pump that we've been discussing isn't going to be adequate, I can return it for something with more capacity, I would just need to know what to order.

Thanks for the continuing discussion!
I usually have a thermostat and timer on my circ line . If I had a wall switch then I would eliminate a timer but still have a thermostat. true woudnt hurt to run pump a little more

#### Wren

##### Member
I usually have a thermostat and timer on my circ line . If I had a wall switch then I would eliminate a timer but still have a thermostat. true woudnt hurt to run pump a little more
What sort of thermostat would you recommend? I could see using one in conjunction with the light switch so that the pump would turn off when the water at the bathroom came up to temperature.

#### John Gayewski

What sort of thermostat would you recommend? I could see using one in conjunction with the light switch so that the pump would turn off when the water at the bathroom came up to temperature.
If you wired a strap on aqua stat to be powered from the light instead of a timer then that should get you what you want. When your looking for am aqua stat there are options as far as water temp and controls to hook up along with the stat.

Replies
0
Views
121
Replies
0
Views
66
Replies
2
Views
118
Replies
7
Views
224
Replies
103
Views
6K
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.