A brick bell prototype to improve standard stove efficiency

[fiedia]

This protoype was built to study behaviours of bells. The purpose is to get data to scale a multibell system. For this experiment a standard wooden stove has been used as heat source..


Stove :
- Godin 3121
- average power : 10kW
- pipe diameter : 125mm (5”)

Bell :
- Bell size : 54cm x 54cm x 112cm (21” x 21” x 44”)
- Single skin of second hand solid bricks (~60x120x240 mm non refractory) mansoned with clay.
- 10mm thick iron cast plate + 3 layers terracotta tiles on top + clay
- steel wool for sealing.


more pictures are available at ecologie-pratique.org/mediagallery/album.php?aid=1847&page=1

Results :
- The bell brings a real heating improvement. Its large radiative area procures a real comfort as soon as the skin temperature reaches 30°C.
- The bell carries on heating even when the stove fire is almost out.
- Wood feeding is much easier. The bell smoothens the draft. The wood will burn even with very little ember left.
- Requires 30 to 50% less wood



To improve :
- Due to the single skin, some smoke leaks through the bell. Double skin at least should be used
- For average fires, the exhaust temperature falls bellow 100°C. Flues get dirty faster. A smaller bell might be better for this stove.
- The bell gets cold too fast (3 to 4 hours). Adding one skin and maing the walls thicker would increase the thermal inertia.

This bell has costed almost nothing but the heating improvement is such that I would advise anyone to try it.

A thermal model has been built to scale such bells.

[smartliketruck]

Unless I missed something, it appears you are feeding the exhaust of a inefficient wood stove into a brick bell.
While this may increase the efficiency in the very short term by pulling more heat from the exhaust, would this not really be a condensation chamber for wood tar/creosote? I hope you have a way of monitoring this very closely.

[satamax]

One spark, and boom!

[peterberg]

First, I would remove the Godin's chimney damper entirely. And run it at the top of its nerves from the moment the bell is added. Don't damp it down, don't try to regulate the burn, that way such a stove would burn surprisingly clean.

And to be honest, a batchrocket could be damped down as well. So the afterburner function won't work and form a lot of creosote on the way. But while the afterburn is happening this is very audible so that will help.

[fiedia]

This type of stove has also a "kind" of afterburn built in. But certainly not very efficient. I very rarely hear it work.

Anyway since I added a bell, I do not need to damp anything (primary air or exhaust). It self regulates. The wood will burn fast but the bell stores the heat. I can wait until there is almost no ember left. The draft induced by the bell will help the fire to restart as soon as I add a log. I do not need to add three logs to get a nice fire. I believe that my fire is now much cleaner since I use exclusively high heat now.

Nevertheless, I explained that the bell is too big for the heat source. The exhaust temp is too low and I have to clean the flue at least every month. If I had to redo it, I would scale it down.

[gadget]

Jun 24, 2018 6:06:53 GMT -8 smartliketruck said:

Unless I missed something, it appears you are feeding the exhaust of a inefficient wood stove into a brick bell.
While this may increase the efficiency in the very short term by pulling more heat from the exhaust, would this not really be a condensation chamber for wood tar/creosote? I hope you have a way of monitoring this very closely.

This topic just came up on permies last couple of days.

permies.com/t/87625/Bell-conventional-woodstove-work

I say give it a go and lets see if there is build up. I'm betting on some wood gas condensates collecting on the inside. Someone needs to be the guinea pig.

-Gadget

[Vortex]

DCish’s bell/bench on woodstove: donkey32.proboards.com/thread/1448/adding-bell-bench-standard-stove

[fiedia]

I am planning on building a second bell prototype

Purpose of this second bell :
- evaluate which bell size is required to heat my living room
- improve a thermal model to scale bells (the goal is to heat 3 rooms with a batch box)
- increase exhaust temperature
- investigate bell efficiency vs height

New features :
- two skins (solid bricks)
- double mass ( ~ 700 kg)
- reduced ISA (internal perimeter 30% smaller, bell csa / pipe csa = 5.16)
- variable height (0.5m to 1.5m by adjustment of internal exhaust pipe length)
- several built in temperature sensors
- inspection / cleaning door
- insulated rear side (aerated concrete)
- concrete top.

Stove in the kitchen :


bell in the living room :


inner skin + roof 1 and 2 :


Bypass + inner pipe with variable length :


stove + bell with its second skin and insulated rear side :


The previous thermal model has been applied to this new bell :


for better image resolution : thermal model

After a two hours high heat fire with this new bell, exhaust temperature of the new bell should be ~ 35°C hotter than the previous bell. Heating power is also reduced by 25 % (1 500 W). But the new bell will heat longer. 9 hours after fire stop, heating power should be improved by 50% (300W for the new bell)

The new bell will have several built in thermocouples :




I will get measurements on :
- smoke temperature inside the bell at different heights. I want to measure the impact of varying inner exhaust pipe height
- temperature gradient through the top, front and rear walls for heat transfer characterization
- living room temperature
- primary air temperature and speed to compute air mass flow
- flue temperature at bell exit, mid point and chimney exit to get the draft

I plan on using an arduino board + PC to monitor about 7 thermocouples in parallel. I will see if the primary air speed is strong enough to use a computer fan as anemometer (commercial ones are quite expensive).

sketchup model

[fiedia]

Some pictures of this new bell building.

It is a provisory set up, so I did not destroy my wooden floor. The bell is put on a multilayer stage : wood fibre board + cast iron plate + quarry tiles. The quarry tile temperature is monitored to avoid floor burning.

Inner skin :
        
Outer skin with 1cm thick cardboard between inner and outer skins.
 
      

Inner skin top


Bell overview. The tops are cast in basic “refractory” mortar. It broke on the weak points around the exhaust holes. The bell still works perfectly, no smoke, no smell.


After some burns, I noticed that the cardboard was limiting the heat transfer between skins. I dismantled partially the outer skin and removed as much cardboard as I could.
The cardboard temperature never reached 200°C during the previous burns. Nevertheless, the cardboard was brown and fragile on the hottest spots. It would have burned sooner or later. For next bells, I will avoid to leave cardboard inside the bell as it limits thermal performances and might be stinky and unsafe
  

There are about 60 thermocouples inside the bell. The setup can read 32 thermocouples at once thanks to 32 max6675 amplifiers, 4 multiplexers and a basic arduino board