Hi Dana,

I was not concerned about the downstairs unit as the ducts are not exposed to a very hot attic. I plan to change it out too as it's very old and uses r22 but I'm going to change it out with a unit but with a heat pump.

I also found out that a ducted style heat pump and a ducted minisplits are essentially the same thing.

I think for now I'll go with a conventional style heat pump and do ductless for when I build my efficient house

Posted By 1201 on 21 Oct 2018 12:14 AM
Hi Dana,

I was not concerned about the downstairs unit as the ducts are not exposed to a very hot attic. I plan to change it out too as it's very old and uses r22 but I'm going to change it out with a unit but with a heat pump.

I also found out that a ducted style heat pump and a ducted minisplits are essentially the same thing.

I think for now I'll go with a conventional style heat pump and do ductless for when I build my efficient house

A primary and very real difference between a "standard" heat pump and a mini-split is modulation. Old school stuff cycles on/off, which is both less efficient AND less comfortable. Old school stuff that's oversized by very much can have problems handling latent loads.

The efficiency differences are sometimes much bigger than the mere SEER & HSPF numbers might imply, largely due to the extremely high efficiency at part load that a mini-split will deliver. Even if they're pretty much the same at maximum speed, sized correctly a mini-split is only in the max-speed range 1% of the time, whereas a 1 stage heat pump can only run at max speed, and a 2 stager will hit it's high range much more frequently, and cycle on/off a lot, taking a hit in efficiency with every spin-up from zero RPM.

What's similar between ducted and ductless-in-a-duct is the inefficiency of the duct work due to leaks and heat loads. If the ductwork is entirely within the conditioned space it matters less except getting the conditioning to the room in which it is needed. Help is coming in the form of smart registers https://www.wired.com/2015/01/brilliant-air-vents-never-knew-needed/ The hitch is the aux blower. There'd be a whole lot of whistling registers through the house if you kept the single-speed blower. What the world needs is a smart vent with its own blower. Second best is low speed continuous circulation through the duct with the ductless on a wired thermostat in the most distant room, the intermediate registers to be adjusted according to need. The hitch here is you'd need a lot of cfm by late aft in Texas. A thermostat in the duct that kicks on a conventional air handler if the ductless is putting cold faster than the duct can deliver?

At any rate, the argument for ductless is the other green, dollars, and the real impediment is contractors who don't like where HVAC is heading.

Closing off registers increases the static pressures and unbalances the duct system. At "typical" pre-IRC 2009 air tightness for houses and sloppy duct design closing off registers would often increase heating & cooling energy use due to higher air handler driven outdoor air infiltration. In a tight house with tight Manual-D duct designs the are some benefits.

In newer construction in the parts of California where Title 24 prescriptive codes are actually enforced (not always the case) or in certified Energy Star houses closing registers can have an overall energy use benefit.

Ducted mini-splits modulating with load will usually be at the lowest blower speed most of the time modulating the coil temp/refrigerant-output to match load, which creates the least static pressure. They only ramp up to max speed on the blower when nearing the max output levels of the compressor.

It's unfortunate that US manufacturers haven't done more with vapor-injection scroll compressors for their "standard" heat pumps. There has been experimental work over the past decade that proves the performance using Copeland vapor injection scroll compressors, but no actual product offerings. One such experimental project report with a Carrier unit lives here: https://www.osti.gov/servlets/purl/1133089

Hi Dana, if I could get a workable ducted minisplit that would work I'm all in, but don't the newer acs have variable speed?

The xv18 says it's variable speed.

Posted By 1201 on 23 Oct 2018 03:54 AM
Hi Dana, if I could get a workable ducted minisplit that would work I'm all in, but don't the newer acs have variable speed?

The xv18 says it's variable speed.

Sure, they're variable, but by how much?

The turn down ratios of XV18 is no more than 2.5:1. (It might only be 2:1, but they're competing against Carrier's Greenspeed, which is 2.5:1). So at min-speed it's output is no lower than 40% of what it is at max speed that's it. The smallest of the series is a 2-tonner, it's bottom modulation level is about the equivalent of a 3/4 ton single-speed. That's fine if your load peak load is 2 tons, but not much modulation range if it's only 1.25 tons, and hardly any benefit at all if your load is 1 ton or less.

That's compared to about 6.5:1 ratio for the 1.5 ton Fujitsu mini-duct unit, which can vary between ~3K & 20K of cooling output at AHRI conditions.

In heating mode it's modulation range may be a bit more. Take a look at some of the numbers in this thread:

https://www.greenbuildingadvisor.com/question/trane-xv18-heat-pump-how-low-does-it-modulate

Posted By Dana1 on 23 Oct 2018 08:24 PM

Posted By 1201 on 23 Oct 2018 03:54 AM
Hi Dana, if I could get a workable ducted minisplit that would work I'm all in, but don't the newer acs have variable speed?

The xv18 says it's variable speed.

Sure, they're variable, but by how much?

The turn down ratios of XV18 is no more than 2.5:1. (It might only be 2:1, but they're competing against Carrier's Greenspeed, which is 2.5:1). So at min-speed it's output is no lower than 40% of what it is at max speed that's it. The smallest of the series is a 2-tonner, it's bottom modulation level is about the equivalent of a 3/4 ton single-speed. That's fine if your load peak load is 2 tons, but not much modulation range if it's only 1.25 tons, and hardly any benefit at all if your load is 1 ton or less.

That's compared to about 6.5:1 ratio for the 1.5 ton Fujitsu mini-duct unit, which can vary between ~3K & 20K of cooling output at AHRI conditions.

In heating mode it's modulation range may be a bit more. Take a look at some of the numbers in this thread:

https://www.greenbuildingadvisor.com/question/trane-xv18-heat-pump-how-low-does-it-modulate

Thank you very much Dana. So ideally a ducted mini split would be more efficient than the normal heat pumps. Now I just have to figure out how to get something like that to work with existing ducting. Sounds like it will be hard. There isn't a lot of info out there in these units or I'm not looking in the right place. Thank you.

You're right that there isn't a lot of general info out there on how to marry them to existing duct designs, but a real duct designer can easily do that math. It's not hard, but somebody has to run the numbers.

Most duct installers are hacking using rules of thumb and get stuff that's sorta-working, with any deficiencies in the design or implementation masked by the oversizing and higher static pressure tolerance of full sized air handlers. But that oversizing of the ducts by the rule-of-thumb guys turns out to be of benefit for right-sizing with the more powerful mini-duct cassettes (primarily Fujitsu). The mini-duct cassette can't deliver the same cfm to those ducts that the big air handler did, but if it's right sized it doesn't have to. At lower cfm the static pressure drop on the duct is lower. Some of the wimpier mini-duct cassettes out there really can't handle much length at all, even when oversized. This too can be determined by a competent duct designer reviewing the system.

Is the existing system in your home built to Texas SOP (separate units and ductwork for each floor)? And does your house sit on an uninsulated slab on grade? The two are related. Between mass effect and convection, the downstairs unit idles, figuratively speaking, through the summer and ditto for the upstairs unit in winter. If the downstairs is doorless and free of unusual heat loads, cooling can be achieved with modest cfm because the slab is distributing it right along with the ductwork.The slab is also contributing given the thermal lag in soil temps under your house. (Peak heat Novemberish, peak cold Aprilish.) Heating is a separate question and, sad to say, a separate duct design.

Which brings me to my point. A hack is likely as good as it gets. My first take is to refit the downstairs AC closet with insulation and a variable speed air handler. You'd want to isolate the ducted mini from the circ air so it doesn't screw up its variable speed functions. One approach is to divide the closet vertically so the air handler sits above the ducted mini in what is effectively a vented plenum. You'd need a wired thermostat and some means of turning up the air handler as the temp falls in the closet. (Or rises.) You'd want to spitball a minimum continuous CFM to keep the mini output moving in the right direction.

A second, simpler approach is to skip the air handler, build a sealed plenum in the top part of the closet and rely on the mini to move air. Your heating load is the potential hitch here. Where I lived in North Texas, I'd opt for an air handler with aux resistance coils. That slab is not your friend in January.