Testing a Prototype Dryer Heat Recovery Heat Exchanger

There has been a discussion going on in the Ecorenovator Forum for a while on figuring out a simple way to recover heat that the dryer is just exhausting out of the house.  I'd been thinking about doing some kind of dryer heat recovery, and the forum discussion got me activated. 

This page covers the results of the first couple tests of the prototype dryer heat recovery heat exchanger I did.  Construction of the prototype is shown here.

I'd like to hear any thoughts on how to improve the design or do it differently.  If you have any ideas, please post on this Ecorenovator thread or in the Comments section below. 

Dryers are very wasteful on the energy front.  A lot of the heat they generate to dry clothes just gets exhausted to the outdoors rather than actually being used to evaporate water in the clothes.  In DOE dryer testing typical US dryers are about 50 to 60% efficient in terms of how much energy they use to dry the clothes compared to the minimum energy needed to evaporate the water.  

Its actually worse than this because as the dryer exhausts heat out of the house, new cold air has to be drawn in somewhere to  replace it the exhausted air, and this new air has to be heated up by your furnace.

The idea of this first cut prototype is to see if its feasible to recover a worthwhile amount of the dryer exhaust heat.

Bottom line is that it looks promising, and there are some potential improvements that could make it a lot better.  Any comments/advice would be appreciated. 

Page Table of Contents

- Description of Heat Exchanger

- Test Load with Heat Recovery Heat Exchanger

- Energy Used and Energy Recovered

- Other Design Issues (pressure drop, lint, water, ...)

- Potential Improvements

- Energy Summary -- Various Schemes

- References

- Comments


Description of the Heat Exchanger

This prototype was made with materials I had on hand -- I'm not recommending these materials for a heat exchanger that will have a long life.  Its more of a get an idea what the performance is and to identify problems prototype.

The heat exchanger is a simple stackup of 14 twinwall polycarbonate sheets that are each spaced apart using spacers at the edges.  Its a counterflow heat exchanger with the dryer air flowing through the spaces between the twinwall sheets in the up direction and the room air flowing through the twinwall sheets in the down direction.

dryer heat recovery heat exchanger
Stackup of twinwall sheets with entry and exit air delivery boxes.
The small boxes are inlets and outlets for the dryer air.
The large box on the top is the room air inlet box with the fan removed.


The picture below shows the heat exchanger mounted on the utility room wall -- the dryer is just on the other side of the wall.  The exhaust duct from the dryer comes in at floor level and goes up to the lower right corner of the heat exchanger.  The duct leaving at the upper left goes outside.

dryer heat recovery heat exchanger

Red arrows  -- show hot dryer air coming into bottom corners of HX.
Blue arrows -- show cooled dryer air leaving HX at upper corners.
Green arrows -- show room air entering HX via the fan, flowing straight down the HX and exiting the bottom (hopefully warmer).

The fan is turned on by the capillary switch mounted to the right of the fan.  The sensor for the switch is the tube you can see in the lower right dryer air entry box with the clear cover.

All the details on the construction here...

Running a Test Load Through the Dryer With Heat Recovery Heat Exchanger

These are the details on running one of the test loads through the dryer,  it covers the results and also asks some questions, and passes on some thoughts on better designs.

Dryer is a Maytag Neptune -- approximately 2001.

Settings were:  Sensor Dry = on,  Dryness = "Very Dry", Temperature = "Regular" (the highest temperature setting).

I realize the "Very Dry" setting uses more energy, but on the other hand some of the items were still a bit damp at the end.

Weight of Water in the Load

The dryer load was several towels, a bath matt, and some odds and ends -- a good sized load.

Wet weight = 17.63 lbs

Dry weight = 9.75 lbs.

Water weight removed = 17.63 - 9.75 = 7.88 lbs

The energy required to evaporate 7.88 lbs of water is (970 BTU/lb)(7.88 lbs) = 7644 BTU or 2.24 KWH
where 970 BTU/lb is the heat of vaporization for water -- the energy required to evaporate 1lb of water.

 So, this is the minimum energy required to evaporate the water this dryer load.

Energy Used and Energy Recovered

This section covers all the energy used to run the dryer and to heat the cold air that the dryer pulls into the house, and also covers the heat recovered by the new heat exchanger -- and discusses some improvements.

The section after this gives a summary of energy use for a regular dryer and the dryer with heat exchanger and a further improved heat exchanger.

Electrical Energy Used

dryer energy use

The dryer fan and dryer tumbler motor use about 260 watts and runs for the full cycle using about 0.26 KWH.

The heating element uses about 5150 watts and ran for most of the drying cycle.

Total power usage was 4.26 KWH for this load.

This was recorded on a TED energy monitor that was just hooked up to the dryer circuit.

HVAC Energy Used

As the dryer air is exhausted outside, new air is pulled into the house to replace the exhausted air.   If its cold or hot outside, then this air must be conditioned by your furnace in the winter or your AC in the summer.   So, this take more energy.

For this load, the outside temperature was 20F, and the amount of heat to condition that air is:

Heat Required = (Flow Rate)(Density of Air)(Tinside - Toutside)(Specific Heat Air) / (Furnace Efficiency)

   = (142 cfm)(58 minutes)(0.065 lb/cf)(70F - 20F)(0.24 BTU/lb-F) / (0.8) = 8030 BTU or 2.35 KWH

Where 0.8 is our furnace efficiency.

So, in this case, it takes half as much energy for the furnace to heat the air that the dryer pulls into the house as it took for the dryer to dry the clothes.

Total Energy Used

So, the total energy used is 4.26 KWH of electrical energy used by dryer plus 2.35 KWH of propane for the furnace.

A total of 6.61 KWH for this load.


Energy Recovered by Heat Exchanger

Heat Exchanger Parameters

The airpath for the room air is through the cells in the twinwall sheets.  The room air comes into the top of the HX, flows straight through the HX, and exits straight out the bottom of the HX.

The dryer air flows in the spaces between the twinwall sheets and in the opposite direction (this is a counter flow heat exchanger).   The hot dryer air flows into the two lower corners (see picture above), through the HX, and then out the two upper corners.  The dryer air was channeled through the spaces between the twinwall to lessen the chance of lint clogging the HX (more on lint below).

Size of each twinwall plate is about 36 inches long by about 21 inches wide (about 20 inches net inside).

Total heat exchanger area (counting top and bottom of each interior twinwall sheet) is (20 *36)(13*2)/144 = 130 sqft.

Flow area each way is about (14 plates)(20 inches wide net)(0.25 inch high) = 70 sq inches  (about 6 times the area of a dryer duct).

It would have been nice to make the plates longer to allow more complete heat exchange, but I did not have the material to do it.  The heat exchanger in the Danish paper is about twice as long, but the graph they have showing the change in efficiency with length indicates that one half as long would still have achieved about 80% efficiency.

Spacers were added between the sheets of twinwall on the end to block the flow of dryer air into the flow path for the room air -- the picture was taken before these were installed.

Room air side of HX:

The room air side flow passages are 1/4 inch tall, and about 5/16 inch wide.  There are 14 layers of twinwall that the room air goes through, and each is 20 net inches wide, and has about 64 of the 5/16 ths wide flow passages.  Total area is 70 sq inches (0.486 sf).

For the test, a short outlet duct was added to the bottom of the heat exchanger that was the full size of the HX -- this was to allow a more accurate air velocity to be measured.

Outlet duct area 7.75 inches by 21.5 inches is 166.6 sq inhes or 1.16 sqft

Velocity was measured with a Kestrel wind meter at 9 locations in the exit duct on a grid.   An Alnor Velometer was also used as a check on velocities.
Average outlet duct velocity = 176.6 fpm

Flow rate for room air through the HX = (176.6 ft/min)(1.16 ft^2) = 205 cfm

The fan that drives the room air side is a Grainger/Dayton 10 inch fan.  It only draws 27 watts, and its power consumption is pretty much ignored in the performance numbers below (27 watt-hours out of 6000+ watt-hours used by the dryer).

Dryer side of HX:

The dryer side flow passages are 1/4 inch tall, and about 20 inches wide.  There are 13 of the dryer flow layers between the 14 sheets of twinwall.  Total area for dryer air is (13)(20)(0.25) = 65 sq inches (0.45 sf).   This is about 5 times the area of the 4 inch dryer duct.

The 4 inch dryer duct has an area of 0.087 sf.

Air velocity in dryer duct was measured with an Alnor Velometer and a  Kestrel wind meter in the straight section of duct between the floor and point just below HX where the duct splits in two.

Inlet duct velocity = 1630 fpm

Flow rate for dryer air through the HX = (1630 ft/min)(0.087 ft^2) = 141.8 cfm

The velocity of the dryer air in the actual HX is  (141.8 ft^3/min) / (0.45 ft^2) = 315 fpm  or 3.6 mph  (is this enough carry droplets of condensation along in the flow?)


Heat Exchanger Performance


dryer heat recovery heat exchanger


The dryer starts at 6:55 PM (the first set of triangles) and the heater element runs continuously for the first 45 minutes until 7:40.  At this point the heater cycles off for a few minutes to keep the temperatures from getting too high, and then comes back on again for a few minutes before the dryer "cool down" starts at 7:48, which lasts until the dryer turns off at 7:55 pm.

Maximum temperature of the dryer exhaust is 153F.

The top red curve is the temperature of the air leaving the dryer, and this gets cooled down to the temperature of the blue dash curve by the time it leaves the HX.  The green dash curve is the temperature of the room air entering the HX, and it gets heated up to the purple dash-dot curve in the HX. 

So, looking at time that that is about average (where the vertical red line is), the dryer flow comes into the HX at 136F, and leaves at 88F for a 48 F drop in temperature.  And, the room air comes into the HX at 79F and leaves at 95F  for a 16 F rise.

The dryer air exits the HX at 88F with a room temperature of 79F, so it taking the dryer air to within 9 F of room temperature.

Part of the reason that the rise in room air is not as large as the drop in dryer air temperature is that the room side air flow rate is about 44% higher.

Heat Taken Out of Dryer Air:

Rate of Heat Removal = (Temp drop)(flow rate)(density)(specific heat of air)

= (136.3F - 88.1F)(141.8 cf/min)(60 min/hr)(0.065 lb/cf)(0.24 BTU/lb-F) = 6397 BTU/hr or 1.87 KWH/hr

If the heat exchanger were perfect, the 136F dryer air would be cooled by 57.5F (down to room temp) rather than the actual drop of 48.2F, so I guess you could say that on a temperature basis is its about 48.2/57.5 = 84% efficient at this point in the cylce-- not so bad for a made out of scrap HX :)

The 0.065 lb/cf for air density is lower than the usual 0.075 lb/cf because of our altitude.

Heat Gained by the Room Air:

Rate of Heat Gain = (Temp drop)(flow rate)(density)(specific heat of air)

= (94.6F - 78.8F)(204 cf/min)(60 min/hr)(0.065 lb/cf)(0.24 BTU/lb-F) = 3031 BTU/hr or 0.89 KWH/hr

It seems like the heat taken out of the dryer air stream should equal the heat gained by the room air stream -- they are the flows on the two sides of the same heat exchanger.

Just guessing at why they are so different: 1) There is some air leakage from the dryer side of the heat exchanger to the room, 2) The heat exchanger surfaces get warm and lose heat to the room directly from its outer surface, 3) the dryer duct into the HX passes right through the room air exhaust from the heat exchanger for about 2 ft, so the hot duct loses some heat to the room air before it gets to the HX, 4) if some of the water vapor in the dryer exhaust is being converted to water, that removes heat from the dryer exhaust side of the HX without recovering it to the room air side? 5) measuring the velocity of each of the two streams is always challenge and there is bound to be some error there.

The first three reasons listed above mean that more heat is actually being recovered than the 3031 BTU/hr -- hard to say how much.

The HX outlet duct velocity is about 20% less than the HX inlet duct velocity, so the first reason in the list above is probably an important one.

For the purposes of the claimed heat recovery that gets to the room, I'm going to say its half way between the 1.87 KWH and the 0.89 KWH, or 1.4 KWH.

Note that on some of the other tests, heat removed from dryer stream is closer to the heat added to room air stream, but there is always a significant difference.

Any ideas on what might be accounting for this pretty large difference?

One thing that would make the HX more efficient is a good steady source of cool air for input.  The heat exchanger is in a utility room that started at 72 F, and got heated to 80 F by the dryer heat exchanger -- it would work better if there was a steady source of (say) 65F air for the HX input.  This would cool the dryer air to a lower temperature, extracting more heat from it.

Other Design Issues

These were some other issues that the prototype was built to explore.

Flow Resistance of Heat Exchanger

One concern was that the heat exchanger would add too much air resistance to the dryer flow.

The dryer duct velocity and flow rate with a very short duct was 1850 fpm and 161 cfm.

Adding the heat exchanger and about 10 ft of ducting plus the flipper vane on the house wall brings the flow rate and velocity down to 1630 fpm and 142 cfm. 

This seem fine to me and is right in the range of flow velocities mentioned as typical in the dryer testing reference. 

So, I think its fair to say that at least this heat exchanger does not add too much flow resistance.


Not a Lot of Liquid Water Visible

I expected to see water in the heat exchanger during the drying process as the high humidity dryer exhaust air was cooled down below its dew point in the heat exchanger and water condensed out.  One of the reasons for building the prototype was to see how much water and where it went and what might be done to make the heat exchanger life OK with the water.

From near the start of test to about the beginning of the dryer "cool down" period, there was some condensation visible on the upper two windows and in about the upper 2/3rds of the twinwall.   Most of this was very small (like a fogged windshield) but, there were a few droplets that got started and were making their way down the twinwall.  I did not see any moisture make it to the bottom of the HX.

HX condensation
Condensation upper dryer air outlet box and
inside the top part of the twinwall stackup.

A few drops of water making there
way down the inside of the HX.

As soon as the dryer cool down period started, all signs of moisture went away.

Using this dew point calculator, if you take an the middle of the dryer run with dryer exhaust at 130F, and assume a relative humidity of 80% for the dryer air, then the dew point is 123F.

Since the dryer air at this time is being cooled to about 88F (which is well below the dew point), it seems like there should be water condensing out -- lots of water?

Perhaps the condensing water is being swept out in the air stream.  The velocity of the dryer air moving up the heat exchanger is 3.6 mph.  

Because of the construction with twinwall, I can see into the heat exchanger and can see the wood fillers that block off the dryer heat exchanger channels at the end -- no visible water.

Any thoughts on where the nearly one gallon of water is going?

From a making life easier on the heat exchanger point of view, the absence of water is nice in that it is easier to build one that would have a good life, but it also means that the heat used to evaporate the water is not being recovered.   See the Improvements section below.

Not Much Lint Either

I intentionally did not put any extra lint filter in the dryer outlet duct just to see how bad it would be and whether there would be any tendency to plug up the heat exchanger passages (which are pretty large). 

After 4 loads, I don't see any sign of lint anywhere in the heat exchanger or in the entry/exit plenums.

Maybe lint takes a while to build up, or some clothes are more lint producers than others and we have not gotten to them, but no lint so far.  There is, of course, a lint screen on the dryer that gets cleaned before each load, and it does accumulate lint.

If we go ahead with a "production" version, I do plan to add an easy change lint filter.

Thermal Images

dryer heat exchanger thermal image

Bright spots on the bottom left and right are the boxes where the hot dryer air enters the HX -- around 116F.

Dark spots to right and left near the top are the boxes where the cooled dryer air leaves the HX -- around 76F.

The hot circle right at the top is the hub of the fan that blows room air through the HX.


What Could be Improved?

Stage 1:

The heat exchanger as it is saves about 32% on the total energy used by the dryer electricity use, and this seems worthwhile.  It could be made better by:

- Pulling air from a cool place and sending the warmed air to another room so that the HX has a continuous supply of cool air.   This does add some complication -- maybe just ducting the outlet air to a place that could use it would be enough?

- Making the HX longer to improve the efficiency (perhaps as much as 10%).

- Try to work out a way to recover liquid water from the dryer airstream to gain back the heat that went into vaporizing it.

- Building it tighter with less leakage.   Although a bit of leakage from the dryer flow into the room does recover some added heat :)

- More durable and water proof materials.


Stage 2:

If the heat exchanger design could be changed such that it condenses out most of the water in the dryer exhaust stream, it seems like lots of benefits would result:

These two additional savings are worth a good deal more than the heat recovery that the heat exchanger achieves now.

What else might be done???  -- got any ideas?


Energy Summary -- Various Schemes

In these graphs:

Electricity-- is the total electricity used by the dryer
HVAC -- is the energy used by your furnace to heat the outside air that the dryer pulls in
Heat Recovery -- is the heat recovered from the dryer exhaust by our heat exchanger
Condensing -- is the energy recovered by condensing water in the dryer exhaust stream

Standard Dryer -- vent outside, no heat recovery (business as usual for most homes):

energy summayr standard dryer

6.65 KWH total


Standard Dryer with Heat Recovery Heat Exchanger

dryer with heat recovery

5.35 KWH total
This is a 32% saving based on dryer electricity, and 21% based on total energy.
This is the configuration used for this test.

Standard Dryer with Water Removal via Heat Recovery HX and Inside Venting

1.3 KWH total
This is a 126% saving based on dryer electricity use, and an 80% based on total energy.



Are We Missing Energy Savings in Clothes Dryers?, Paul Bendt

New counter flow heat exchanger designed for ventilation systems in cold climates, J. Kragh *, J. Rose, T.R. Nielsen, S. Svendsen




blog comments powered by Disqus