Newbie duct “designer” seeking practical experienced advice.
kermit49
| Posted in Mechanicals on
Hi guys, my very first post here.
I am 73yo designing our retirement home for wife and I. Zone 3 southern Utah, 2100 sqft, one level. Manual J load calcs 14,850 heating, 16,730 cooling. House to be as tight and well insulated as practical (R27wall/60attic), all electric, HRV, everything in conditioned space.
There will be one wall mounted mini covering all open areas, about half the house load. No issues here.
The second mini will be a ducted Fujitsu 9k servicing four bedrooms and 2 baths. Obviously using doors.
These ducted rooms are getting their required cfm (per heat load calcs). All duct and fitting velocities run close to 400 fpm target +/- 30 cfm. All duct and fitting losses are calculated per formulas. Ducts are all steel except for last 2 foot leading to each diffuser which is stretched flexible duct.
Thanks to one our own contributor MattJF, I downloaded the McGill Airflow design pdf he recommended. What a great tutorial ! Highly recommended.
Contents.PDF (mcgillairflow.co
Contents.PDF (mcgillairflow.co
Used, what I later realized, is the “constant velocity” method. I also learned to calculate system static pressure. Sys Static = Critical Path Supply losses – Fan output velocity press. The critical path losses (supply side only), from plenum to furthest diffuser, totaled 0.100 iwc.
The system fan output velocity pressure of 0.007 iwc was obtained from this Ultratech chart
The resulting System Static press = 0.100 – 0.007 = 0.093 iwc.
MY ASK :
Never having done HVAC design, I would appreciate having a reality check on my ducting design from those that have ducting system design experience.
1. The Fujitsu 9k at highest rated flow delivers 0.36 iwc static. Calculated, on supply side only, system static press is 0.093 iwc. IS THIS TOO GOOD TO BE TRUE ? Did I overlook something ?
2. Do I need to add the RETURN’S critical path losses to the SUPPLY critical path losses when making the total critical path pressure calculations above ?
3. The McGill Aiflow design pdf did not explain the effect adding the return side components would have, if any. However, Manual D’s example did include the return side components (using equivalent length method) when calculating Friction Rate. Huh!
Do duct system designers use the return critical path components when calculating over all system pressure losses ?
4. Return filters. Does each return branch discharge into a return side plenum feeding a sizeable filter leading to the Fijitsu ?
OR does every return duct have a filter at room’s collection point? Are either of these two methods ok in actual practice ?
Thank you in advance.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Kermit49,
I'll give you a bump.
Thanks.
I'm assuming you meant 400FPM. At that low velocity, any reasonable ducting will be very low loss. Your ~0.1" is definitely reasonable, I've gotten about half that on installed systems that use very low static air handlers.
The air return and filter adds to your pressure budget exactly the same way as the supply ducting. There is no difference here. For low static air handlers, you have to watch this as well thus the recommendations of very large filters for most ducted mini splits.
If your return ducting is well sealed, you can do filters at each return only. A well sealed return and good filters means both your ducting and air handler will be pristine for many years to come. Replacing four filters is not fun, so I would instead use a larger central filter and maybe a washable filter mesh at each return.
A simpler option is a single hallway return and door undercuts or transfer ducts from each bedroom to the hallway. I've had good luck with converting the ducted unit to bottom intake and mounting a larger filter grill right against the bottom side of the unit. No extra return ducting is needed but it is a bit louder on high speed which is usually fine in a hallway.
Akos, I've have read your writings here over a period of time and have always been impressed. Not surprised at all now reading your good advice and recommendations. Thank you.
1. Yes 400 fpm, hey I am a senior and entitled ; > )
2. You made my day. Akos's blessing on my calcs carries a lot of weight. What a relief.
3. So my pressure budget includes all components in the return path. Good to know before I continue with the design. Would have been a bad learning experience had I not included that. Remember, all I know is little bits and pieces I read here and there, nothing comprehensive. You filled in the blanks for me.
4. Got the big filter message from you on your previous posts, good reminder.
5. Although doable, I would hate having to run another set of metal returns. If I did, yes they will be well sealed especially if I used individual terminal filters. Like sucking a milkshake through straw with a bad split.
6. I prefer a hallway return and would design accordingly, (using room returns to hallway, etc) however, is this common in the mini-split world? Most all my research here tells me when dealing with rooms/doors to duct back to mini-split. Not looking for extra work but.
As mentioned, half the house is open and serviced by a mini-wall unit, no return needed as the hallways will take care of that. So Akos, here is my burning question. If the ducted return relies on hallway air using a hallway grill that lead to ducted mini, any chance the combined return (ducted and ductless) cause control problems on the ducted mini? I think I am overthinking this but with zero experience in these matters I just don't know. My gut feeling is that this set up will work fine.
Akos, thank you for your solid advice, I now know I am not wandering off the correct path.
For the ducted mini splits I've always done very simple returns from hall/open area. About the only time you need to do more complicated return ducting is in a multi story structure where the unit is in the basement.
Around me, door undercuts are the norm for forced air setups, this also works well enough for a ducted mini split in a high efficiency structure. Typical bedroom heating loads, thus airflow, are low enough that a standard size door undercut won't be restrictive. Undercuts are not the best for privacy, if you care about noise, high/low transfer duct inside the wall is a better option, size so that pressure loss through it at your design flow rate is under 5PA.
Most (all?) ducted units come with a wired thermostat. Somewhere on the thermostat or in the install menu you can set it to use the temperature sensor in the thermostat instead of the one in the intake of the unit. If you mount the thermostat inside the main bedroom, it will only pick up the temperature there and control based on that. This way it won't matter if there is an extra mini split in the living area nearby.
Glad to help out, designing a house is not simple and the little details can make a big difference in durability, efficiency and comfort. Good luck with your build.
P.S. If in your climate you have a hard time keeping winter time humidity high enough in the house, an ERV might be a better choice.
Akos, your words are a game changer. I trust your judgement and expertise.
Do I prefer the hallway return method? Heck yes!
Will wrap up the ducted return design I started, for completeness, and start over with the hallway return method.
Been considering this method for some time, researching how to balance each room with wall efficient air flow grills, connecting using sound transmission reducing materials. Room to hallway bridge will not leave conditioned space. Undercuts won't pass wife test. We had that conversation some time ago while researching hallway style return.
Good to know thermostat control won't be an issue. Whew! That's an area I struggled with. Solution is clear now, thank you.
You make a good point on the ERV. Yes our retirement home will see low humidity (Mohave Desert), practically all year but worse in winter. I am drawing all ventilation lines run now even though the purchase of the E/HRV won't happen for a while. Re-routing badly needed moisture back into the house makes sense. In Florisa, seems that re-routing drier air, than outside, back in the house makes perfect sense. Wonder what climate doesn't benefit from an ERV ?
Thank you for the well wishes Akos. Our first new home build. Wish we had done this twenty years ago.
Appreciate your thoughts and real world input.
FYI the link seemed to have changed to this:
https://www.mcgillairflow.com/pdf/productlit/ductSysDesign_guide.pdf
You are correct Dan. Thank you for finding the error. I just corrected it.
It worked a few weeks ago when I downloaded. Thanks again.
Hello Kermit49,
I am in the same boat. Using fujitsu slim duct 9k units to support upstairs bed rooms. How do you calculate "critical path supply losses" using constant velocity 400 FPM? The Manual D seems based on friction rate design. Appreciate you attention.
Regards,Rajib
Hello Rajib, as I said, I am quite new to all this. I trust someone will chime in if my method is lacking. I will give you a high-level response but can add details later if needed.
I picked 400 fpm as a reasonable target to keep sound and register air blasts in check especially in bedrooms.
First check out the Fujitsu capacity tables for the 9k. I wanted to use the 9k but it was short on cooling capacity for my area during the summer season. Too bad as it would have been perfect. I selected the 12k for that reason.
I assume you did a manual J and know heat load for each room to be serviced. Those load numbers will translate into required heating and cooling room cfms. I recently asked about assigning cfms based on room heat load here https://www.greenbuildingadvisor.com/question/guidance-calculating-room-cfm-requirements-based-on-heat-load#comment-235219?utm_medium=email&utm_source=notification&utm_campaign=comment_notification&utm_content=view
For each room, I picked the higher of the two cfms (heating and cooling) as the room design cfm. Adding all room cfms will tell you if the 9k fan has the cfm capacity.
Lay out your ducting on paper from plenum to furthest register. Unless there is another register with high air demand, that will likely be your critical path. You will check other registers later to make sure, right?
What I did was to size fittings and ducts close to 400 cfm selecting available duct sizes. Some flows were higher some lower using off the shelf diameters in increments of 1 inch starting with 3 inches.
After passing a serviced room with X cfm requirements, I then reduce the branch duct size to match new lower flow. I kept doing this throughout.
This seemed simple and logical. Only con will be sourcing fittings that have mixed diameters. Alternative would be to reduce after most fittings.
Friction rate (FR)
I calculated the FR using estimated total equivalent length (TEL) on the supply & return critical path. See Manual D Figure 7-6. This was used calculating duct and any fitting with an assigned TEL. Where I could, I used coefficient factor for fittings such as wyes, elbows and tees.
Got these coefficients from old ASHRAE fitting table. Should be elsewhere.
Register pressure drops from Hart & Cooley. Flex duct pressure drop estimated 3 x metal duct drop. In my case negligible as these are no more than two feet long. If longer I would calculate more precisely.
Adding all the critical path losses on the supply and return critical path gave me total system pressure drop. Now calculate system static pressure.
Sys Static = Critical Path Supply losses – Fan output velocity press (immediately after fan)
In my case the Total Pressure losses totaled 0.100 iwc.
My system fan output velocity pressure of 0.007 iwc was obtained from this Ultratech chart
https://www.kele.com/Catalog/07%20Flow/PDFs/Airflow%20velocity-pressure%20conversion%20chart.pdf
The resulting System Static press = 0.100 – 0.007 = 0.093 iwc. Now compare this to the 9k's static pressure at the cfm mode it will be programmed to run, there are 4 cfm choices. Quiet to Maximum mode.
This Rajib this is how I interpreted what is required.