Casting a batch rocket for use as a heat source in an outdoor pizza oven/parilla

Following on from a thread started by Matteo Rossi (https://permies.com/u/366267/Matteo-Rossi) it was suggested that I start a new thread with my experience of casting a batch rocket.

So here goes...

First let me provide a bit of background:

This batch rocket is for providing a heat source for a pizza oven/parilla.  It is positioned at ground level but is thermally isolated (as much as is possible) from the floor and block structure that surrounds it.

The pizza oven is heated via conduction from the top of the batch rocket and the riser. The batch rocket is surrounded by loose fill perlite and underlain by insulating firebrick except for a steel plate that sits on top of the batch rocket chamber.

Here's a design picture.


The batch rocket design itself came courtesy of Peter Van Der Berg and Sketchup.

Very interesting design.
What size will be your batch box, 4", 6"?
Will the oven be black (with fumes going through it) or white?
Have you made a prototype? I'm asking, because I think that if the oven in this configuration is white it will not reach baking temperatures. If it did it would be great.

Hi Stephen,, welcome to the forum.
I have often thought a batch box pizza oven  hybrid would be a good design but as Cristobal points out, you need a lot of BTUs to get a brick pizza oven up to temperature.
I would like to see your idea in more detail?

Perhaps a design where you can still light a fire inside the brick dome and utilise the rocket stove to maintain the  temprature of a pre heated oven or, re heat the next morning for a day of slow cooking?

If your design is a white oven, the lower floor bricks, after a few hours of burning, may well reach the required 360- 400c but the  upper dome bricks are unlikely to reach wood fired  oven temratures. The top of the dome in a well built pizza oven will be 500c.
That is not to say the oven wont work as a more conventional oven at lower temratures.
Other folk have made rocket stove pizza ovens from metal drums using radiant heat rather than heat saturated bricks.
This style of oven can be found on the forum and the users are happy enough with the results  but, honestly a conventional high mass pizza oven is a specialist  piece of cooking equipment, designed to do the job  and developed over hundreds of years.
You may be interested in Peters new designs where the riser is much shorter or even ‘no riser at all’
One of these  riser less designs used as a black oven might be worth looking into.
It looks like you have molded you batchbox core in one piece, can you give a few more details about how you built it?

Cristobal Cristo wrote:Very interesting design.
What size will be your batch box, 4", 6"?
Will the oven be black (with fumes going through it) or white?
Have you made a prototype? I'm asking, because I think that if the oven in this configuration is white it will not reach baking temperatures. If it did it would be great.

Batch box is 150mm diameter riser which I think is the dimension you are referring to.
Oven is planned to be white... but in the event it doesn't get hot enough - I will have to go black.

You make an excellent point about baking temperature hence my Plan B. However I've tried to mitigate this problem by a design that surrounds the batch rocket in loose perlite fill and mounts it on insulating brick. The logic being that if I can reduce the rate of heat loss around the batch  rocket it will flow into the space above. The proof of this will be in the pudding!!

Will keep you informed of progress.

Regards

Stephen

Fox James wrote:
Hi Stephen,, welcome to the forum.
I have often thought a batch box pizza oven  hybrid would be a good design but as Cristobal points out, you need a lot of BTUs to get a brick pizza oven up to temperature.
I would like to see your idea in more detail?

Perhaps a design where you can still light a fire inside the brick dome and utilise the rocket stove to maintain the  temprature of a pre heated oven or, re heat the next morning for a day of slow cooking?

If your design is a white oven, the lower floor bricks, after a few hours of burning, may well reach the required 360- 400c but the  upper dome bricks are unlikely to reach wood fired  oven temratures. The top of the dome in a well built pizza oven will be 500c.
That is not to say the oven wont work as a more conventional oven at lower temratures.
Other folk have made rocket stove pizza ovens from metal drums using radiant heat rather than heat saturated bricks.
This style of oven can be found on the forum and the users are happy enough with the results  but, honestly a conventional high mass pizza oven is a specialist  piece of cooking equipment, designed to do the job  and developed over hundreds of years.
You may be interested in Peters new designs where the riser is much shorter or even ‘no riser at all’
One of these  riser less designs used as a black oven might be worth looking into.
It looks like you have molded you batchbox core in one piece, can you give a few more details about how you built it?

Agreed re: energy/heat required to get oven up to cooking temperature. However with the insulation I have in place I'm hoping that the heat will flow into the steel plate on top of the rocket and into the oven space. Also I'm hoping that by coupling the riser to the roof I will get conduction there as well.

You mention a high mass oven. It's important to recognise this is not a high mass oven. It's a low mass oven. The only mass in the oven is the refractory arch. Behind that is insulation. Around the rocket oven is insulation.

You mention Peter's new design which I will have to look at. My plan C is to remove the riser so that the oven turns into a black oven. So thanks for the tip regarding Peter's design.

I will be documenting everything, but still just getting to grips with the forum's formatting quirks.

Regards

Stephen

Here's my first installment of the build...

I found it easier to write a document then attach a pdf rather than use this editing tool.

The batch casting...

Let's see if i can do the next part using this forum editor.

Riser mix and geometry
Now onto the riser... I used a homemade refractory mix for this rather than the commercial refractory that I bought for the batch rocket.
I used a standard mix of 3 parts cement, 3 parts perlite, 4 parts fireclay, 4 parts silica sand and 4 parts water BY VOLUME!
Courtesy of http://backyardmetalcasting.com/refractories.html and Lionel Oliver.


Calculating volumes is trivial if you use metric and a PITA in hobbit feet and cubic fingers.
Dext: external diameter: 210mm. Dint: internal diameter: 150mm. H = 145mm
Mould volume in m3 and L is 3.14 x 0.145 x (0.105^2 – 0.075^2) ~= 0.0025 m3 = 2.5L
Using a material volume of (1.5 x mould volume) I can then calculate what the mix should be. The coefficient of 1.5 accounts for the fact that the materials fill their relevant spaces between grains.
Material volume estimate = 2.5 L x 1.5 = 3.75 L
Sum of parts of materials (3+3+4+4) = 14.
• Cement volume required:  3.75 x 3/14 = 0.8 L
• Perlite volume required: :  3.75 x 3/14 = 0.8 L
• Fireclay volume required: 3.75 x 4/14 = 1.07 L
• Sand volume required: 3.75 x 4/14 = 1.07 L

The calculation is a bit trickier in hobbit feet:
Dext = 8 1/4”, Dint = 5 7/8”, H=5 5/8”... This gives about 264 cubic fingers... which is about 0.027 of a US dry barrel and about 4.5 x 10-6 of a cord. If you want that in Gallons you’d best speak to Harry Potter or check with Charles Dickens. Anyhow back to the 21st century.

Here’s a sketch of the riser mould. This was made with wooden slats on both the inside and outside of the riser.
I used an 8 piece arrangement with the table saw set at 11.25 degrees to bevel  each slat such that 8 pieces form a ½ cylinder.

Now let's try to embed an image or few...

I can't seem to do anything except have the images bunch up at the end of the message.

I hope they make sense without much text by way of explanation.

Regards

Stephen

Stephen,

I hope you will not have to use plan C, because the beauty of your design is to have the cooking plate AND the oven.
Doing one at a time is totally doable, but both not so easy.

On the other hand I have noticed in my primitive cob oven that no matter how much I heat the dome, if the floor is cold the pizza or bread will remain too wet at the bottom and the top will be charred. I have no insulation on the floor which is just half a ton of mortared CEB block base. So I started leaving hot coals on the floor and bake pizza on cast iron plate laid on the coals. 5 minutes was sufficient for perfect bake. After that I realized I would rather have hot floor and cooler dome than the opposite.

Regarding the "shoe-box" batch box without the vertical riser, someone on donkey32 forum has built one with intention to heat the cooking plate and it turned out to be a failure, but it does not mean it could not be improved. I think the temperature of exhaust from shoe-box is lower than from vertical riser.

Cristobal

I hope you will not have to use plan C, because the beauty of your design is to have the cooking plate AND the oven.
Doing one at a time is totally doable, but both not so easy.  

Indeed.

I've made the opening to the oven wide enough so that I can remove the steel plate in the oven and replace it with an insulating sheet, which would then cause more heat to flow up into the riser and toward the cook plate on the parilla.

If I can alter the height of the cook plate on the parilla above the riser, I'm hoping that might allow me to adjust the heat into the plate without damaging the all important process of efficient pyrolysis down at the base of the riser.

Someone asked earlier in the thread if I'd made a prototype and the answer is no. This is the rather expensive prototype that I'm hoping I can adjust as I move forward depending on performance.Presently, I've cast the riser and the batch rocket. The pizza surround is up to the 3rd level of block and once I've finished testing the batch rocket I will install that in position within the structure and start the next phase of construction: the oven arch.

My concern with the oven is that I'll end up with burnt bottoms and wet toppings in my pizza. One idea to mitigate this would be to have a riser with a few holes in it... to create a kind of "grey oven".

Anyone ever heard of that idea?  Does it bugger up the efficient rocket oven pyrolysis?

All fun and games...

Regards

Stephen

Stephen,

How about making the arch from steel? You would get the conduction heating of the floor and convection of the riser exhaust would heat the steel dome.

That's a very interesting point...  Problem is i've already cast 60 refractory bricks for the arch!

Following up on the heat issue, I've just done a quick back of a fag packet calculation.

The specific heat capacity of refractory brick is c. 1000 J/kg.C. If I have 60 bricks each weighing about 2kg. I’ll need 1000 x 180 Joules to raise the temperature of the brick by 1 degree = 180 kJ/°C.
If we needed to get this up to 400 °C, that’s 400 x 180 kJ = 72 MJ. Wood has roughly 15 MJ/kg. So even with zero energy loss and 100% efficiency I’m going to need 4.8 kg (72/15) kg of wood.

Hmmm... wish I’d done this calculation before!

If I can keep heat loss down to say 50% of the heat required... which I doubt, that’s 9.6kg of wood. A seasoned medium density wood is going to require 15 L of wood (that's roughly a bucket load for our imperial friends).

This is beginning to look like it's soggy toppings, without some blackness in the oven...

You summed it up nicely. In case of heating from the inside, the same brick may not need so much energy, because the interior side of the brick will reach desired temperature, before the exterior is heated. it's a main reason why white ovens are sub-par. You have used perlite in your refractory cement which will also affect the heat transfer to the floor of the oven. I would use HD firebricks and steel dome.
You still have the option to make it "grey" which would be really black, but with the loss of heating power of the top plates unless you use a cast iron door: open = heat the oven, closed = heat the plate. However, cast iron may be too delicate thermally to work there, so I would rather use a brick (Or two) to close the heating canal (reaching through the oven).

Hi Cristobal

I should clarify. The base of the oven will be a steel plate which should transfer heat rapidly from the top of the batch rocket into the oven space. Hence my concern about burnt bottoms and soggy toppings. (In retrospect I can see your idea of the opposite might have been more sensible as regards heat flow - ie. steel arch/refractory base).

To get more heat into the oven space - i've also considered making a "skinny" riser - one that is thinner than the usual one and hence heats up and conducts heat more quickly into the oven space.

I've yet to see any of the ovens posted at permies really insulate the batch rocket really "tightly". My hope has always been that by cutting the flow of heat to the base and sides, heat flow will be significant up into the oven. Have you ever seen anyone try this? ... (ie not just a thin insulating layer but thick 200mm of insulation?)

Tomorrow I'm hoping to carry out another burn test in the batch rocket, this time with a proper sealing door and taller riser in place. This will properly test the batch rocket for performance. But I won't be able to test for heat flow until it is surrounded by perlite with the ceiling and floor in place.  That won't be until the end of the month at best.

I have never seen a hard firebrick riser with more than 100 mm of insulation.
I would not worry about the possibility of burnt bottoms. I put my pizza on the cast iron plate directly on hot, bright red coals and the bottom is NEVER burnt. The top could be if the dome was too hot. If the bottom was burnt it would mean the top was solid charcoal. In my experience - the very hot bottom is the key to perfect pizza and even more for bread. The thick dough of the bread needs constant heat coming from the bottom to bake it thoroughly.

In an ideal world you would have 320-340 c on the surface of a porous surface, that will avoid soggy bottoms, the top of the dome should be around 400c.
However a lot depends on what you are looking to cook and your expectations, a big fat pizza with lots of toppings is better off being cooked at lower temperatures and a thin margarita is better at slightly higher temps.
Steel is not a good surface to cook on, at least not for pizza but of course you can do it.
I would guess that a 6” batch box built to spec would heat a steel plate above the riser to around 600 -700c and it will be glowing red hot.
My tiny 4” vortex riser-less core heats the cook plate to around 350c
My 6” vortex J tube has a big hot plate and it will heat the plate red directly above the riser but only 6” away the plate will be around 250c.
Getting back to my first question, did you cast your core in one piece?

stephen wilson wrote:Following up on the heat issue, I've just done a quick back of a fag packet calculation.

The specific heat capacity of refractory brick is c. 1000 J/kg.C. If I have 60 bricks each weighing about 2kg. I’ll need 1000 x 180 Joules to raise the temperature of the brick by 1 degree = 180 kJ/°C.
If we needed to get this up to 400 °C, that’s 400 x 180 kJ = 72 MJ. Wood has roughly 15 MJ/kg. So even with zero energy loss and 100% efficiency I’m going to need 4.8 kg (72/15) kg of wood.

Hmmm... wish I’d done this calculation before!

If I can keep heat loss down to say 50% of the heat required... which I doubt, that’s 9.6kg of wood. A seasoned medium density wood is going to require 15 L of wood (that's roughly a bucket load for our imperial friends).

This is beginning to look like it's soggy toppings, without some blackness in the oven...

I am not disputing your numbers at all, but would be interested in knowing ( perhaps someone has one) the amount of wood required to heat up the pizza oven that uses an internal fire, in the oven it self (such as the photo) I know this is different, your proposed white oven vs the picture black oven type.  But curious none the less.

Well I have a pizza oven and I have been building them for a living during the last 20 years of my life.
I really dont know the weight of wood I use to fire my own oven oven but it takes about one and a half hours to saturate the dome mass and around two days to cool back down.

My ovens are made from cast 2” thick refractory cement and use 4” of ceramic fibre insulation over the dome.
The floors are made from 3” thick fire brick and also have 4” of ceramic board underneath.
I light a stack of wood in the centre of the floor and reload four times to get around 500c in the top of the dome and 400c on the the bricks.

I can cook breakfast in the following morning and slow cook meat in the day.
If I need to re fire the next day it takes around 45 minutes to get it back up to 500c.
The time of year and the wind can have an effect but those are average factors.

I have a riser-less vortex rocket stove with a large hot plate it operates at 750c in the afterburner and the hot plate is around 250-300c for cooking on.
My vortex J tube has a big hotplate and runs around 500c in the centre going down to 150 around the edge.

I would guess a 6” batch box built to spec would heat a hotplate above the riser to around 600c or red hot in the centre.
Metal has the habit of burning pizza bases at wood oven temps……

Fox: "Getting back to my first question, did you cast your core in one piece?"

No, I cast it in 2 halves. With the riser elements also cast in halves.

Any recommendation on the best way to join them?

Scott Weinberg wrote:

stephen wilson wrote:Following up on the heat issue, I've just done a quick back of a fag packet calculation.

The specific heat capacity of refractory brick is c. 1000 J/kg.C. If I have 60 bricks each weighing about 2kg. I’ll need 1000 x 180 Joules to raise the temperature of the brick by 1 degree = 180 kJ/°C.
If we needed to get this up to 400 °C, that’s 400 x 180 kJ = 72 MJ. Wood has roughly 15 MJ/kg. So even with zero energy loss and 100% efficiency I’m going to need 4.8 kg (72/15) kg of wood.

Hmmm... wish I’d done this calculation before!

If I can keep heat loss down to say 50% of the heat required... which I doubt, that’s 9.6kg of wood. A seasoned medium density wood is going to require 15 L of wood (that's roughly a bucket load for our imperial friends).

This is beginning to look like it's soggy toppings, without some blackness in the oven...

I am not disputing your numbers at all, but would be interested in knowing ( perhaps someone has one) the amount of wood required to heat up the pizza oven that uses an internal fire, in the oven it self (such as the photo) I know this is different, your proposed white oven vs the picture black oven type.  But curious none the less.

Hi Scott ... I appreciate the curiosity... once I get to fire her up inside the insulated space, I  will let everyone know.

Fox James wrote:

My ovens are made from cast 2” thick refractory cement and use 4” of ceramic fibre insulation over the dome.
The floors are made from 3” thick fire brick and also have 4” of ceramic board underneath.

Hi Fox,

For comparison I have the batch oven sitting on 2 insulating bricks (Lamda = c.0.2W/m.k, U= c. 1.75 W/m2.k) - with the 114mm dimension vertical - one at the front and one at the back.
Underneath the batch oven is about 100mm of loose perlite  (Lamda = c.0.08W/m.k, U= 0.8 W/m2.k)
Then the base is a 100mm thick perlite/sand/cement "concrete" base (Lamda: c. 0.5 W/m.k, U=5W/m2k).

Around the oven is a perlite "shell" (ie wall). Behind that is around 75mm of more loose perlite.

I'll send a section later today. From what you have said I'm thinking i might have under-insulated the base.

Not  necessarily, my ovens are designed to hold heat for long periods and I do over spec the insulation!
You can join the halves with steel wire, my pizza ovens use 100mm long stainless staples to hold the segments together but allow for expansion.
I think you could just wrap some wire around, maybe 1.5-2mm stainless would do it and still expand ok.
Another alternative might be to build a box around  the core and back fill with loose vermiculite, maybe even find a oversize metal tube to surround the riser and back fill with vermiculite. (Perlite might work)

Here's a vertical section through the whole structure, facing the front.

At the level of the batch rocket, the spaces between the inner insulating wall and the middle insulating wall, plus the space between the middle insulating wall and the outer concrete structural wall, will be back filled with loose perlite.


...

stephen wilson wrote:Here's a vertical section through the whole structure, facing the front.

...

Hello Stephen Wilson, I am smiling a bit, as I was told I could never make such a shape with Fire Brick, but it looks like you have, How did you cut the tapers in your fire brick to make such shapes..   I am liking it.

Width of sides?
Height of dome?
Depth of dome?  (front to back)

Cheers
Scott

Here's a picture of the firebrick mould that i made...

It consists of a slab of wood about 250mm wide and about 1m long.

It has small rebates cut in it about 15mm deep/wide into which i can fit plastic "tablilla" ... these are pieces of offcut plastic ceiling tongue and groove. These moulds give me rectinlinear bricks.

If I need bricks for the arches I use angled separators which you can see in the foto with a defined angle on them. The separator that i'm removing in the foto is a 6 degree separator to give me a 6 degree brick.

Scott Weinberg wrote:

Width of sides?
Height of dome?
Depth of dome?  (front to back)

Cheers
Scott

Scott,

As regards your question on size... presently my design plan has dimensions for Height, Width,Depth of:

286mm Height
993mm Width
805mm Depth

However as we all know, plans are made for changing and I'm going to reduce that width down by about 100mm.

The key design parameter is that I'm using a catenary curve for the arch.

Regards

Stephen

Fox James wrote:Not  necessarily, my ovens are designed to hold heat for long periods and I do over spec the insulation!
You can join the halves with steel wire, my pizza ovens use 100mm long stainless staples to hold the segments together but allow for expansion.
I think you could just wrap some wire around, maybe 1.5-2mm stainless would do it and still expand ok.
Another alternative might be to build a box around  the core and back fill with loose vermiculite, maybe even find a oversize metal tube to surround the riser and back fill with vermiculite. (Perlite might work)

Thanks for the Tip Fox.

On another note... don't suppose you could recommend an insulating perlite mix recipe for a riser. I was thinking 3 Perlite, 1 Ciment, 1 Sand, 1 Fireclay,,,, or maybe i can go even lighter with say 4, 1,1,1?

I would like to make the mix fairly light because I've got to put this up through the roof. I will be using a 150mm cardboard tube for the inner part of the mould, and have welded up a 230mm diameter steel tube from 1mm steel sheet.

Regards

Stephen

Scott Weinberg wrote:

stephen wilson wrote:Here's a vertical section through the whole structure, facing the front.

...

Hello Stephen Wilson, I am smiling a bit, as I was told I could never make such a shape with Fire Brick, but it looks like you have, How did you cut the tapers in your fire brick to make such shapes..   I am liking it.

Width of sides?
Height of dome?
Depth of dome?  (front to back)

Cheers
Scott

Thanks Stephen,  this is good to know, I thought you might have had some special saw that could cut 4.5" through a fire brick, my quest has always been, to build this with simple layups and my "least cost fire bricks"  so it is good to know how your doing it.
Best of success

Stephen, how did this oven turn out?

Good morning everyone... sorry its been so long since my last post.

William asked me how it all went in the end...

Stephen, how did this oven turn out?

Pretty damn well is the answer... See figure 1 at the end. The batch rocket works well with no smoke except when lighting - the smoke only lasts a short while before the batch rocket really starts to roar and then all smoke disappears...

I've now had it running now around 10 times... I've been learning how it performs:

Performance
The parrilla works a treat with a very hot zone above the batch rocket which then reduces in heat towards the front where it is not uncomfortable to stand. I tend to put pots on the super hot part for boiling things. In the middle I sear steak/tuna, at the front I cook eggs etc. See Figure 2. This redirects the exhaust gases from the burn tube around to the front of the hot plate before they pass down the sides and then finally up the chimney. There are baffles in the parrilla to make this work.

The pizza oven works well. I load it with one batch to get up to heat, then start cooking pizza on the second load. The door remains closed except when I remove or put in new pizza. Each pizza takes about 10 minutes with a turn half way through. i can cook 2 pizzas at a time. See Figure 4
For a 2 pizza cook and a dog-food cook-up (I have a very lucky dog)... I boil up 2 chickens in a pot with rice and veggies... which creates about 15L of stew... this uses 2 batches of wood
When hosting a large group of people (say 12 guests x 4 pizzas plus steaks etc on parrilla) I get through maybe 4 batches of wood... say 60L.
With a ventilation fan going everyone can sit in the outdoor kitchen without it being uncomfortable - if this was in the UK, I would guess that you'd not need the fan
The front and tops of the oven get warm to the touch but nothing more than maybe skin temperature

In terms of thoughts and modifications,
I suspect that my secondary air pipes will need replacement within a year or two. See Figure 5
Keeping the heat in is critical to pizza temperatures, making a better seal would be a sensible task
Finding a way to transmit/store more of the heat into the pizza chamber might be a good idea perhaps with an extra turn in the burn tube before it exits the pizza chamber
Having thicker insulation under the floor might be a good idea. Presently it has 4" of EPS then perlite concrete base then some loose fill.
Finding a low cost supplier of bulk perlite is critical... just buying from the garden centre is way too expensive

Tasks still to do:

Put handle and insulation on "secondary" batch rocket door
Find an oven thermometer that works
Cast a chimney for the back of the parilla

Comment on permie member concerns:

Getting to a high enough temperature:  I don't have temperature data yet, but the fact that the pizza is delicious is the best data point that I could have. The objective was to get the pizza oven up to temperature. Rather than make it a dirty black oven I was hoping I could use other design concepts to reach that objective. These concepts were: (i) insulating the hell out of all parts of the oven (ii) trying to minimise any thermal bridging  (iii) minimise hot air leaks from the pizza oven (iv) allow the hottest "exterior" parts of the batch rocket to heat the air in the oven (this concept arose after some advice from Fox James ... so many thanks to him). To be clear on that last point the hottest part of the batch oven is at the back, sides and top of the burn tube. These parts are exposed to air that can freely circulate up into the pizza oven. The batch rocket itself sits on 2 thermal insulating bricks. The front of the batch rocket and underneath is partially blocked off with loose perlite. So in a thermal sense the batch rocket kind of sits inside the pizza oven.

Losing too much heat up the burn tube: By swapping out a perlite insulating tube for a piece of well pipe the hottest gases coming out of the batch rocket now passes some of their heat into this thick cast iron tube which sits inside the pizza oven. At the top of this tube is an insulating piece of perlite to minimise heat loss via conduction from this tube up to the parilla. I figured the parilla would get hot enough with just the exhaust gases.

I will post more pictures and provide temperature data as soon as I have some.

One other point... here in Costa Rica, temperatures don't drop below 20 degrees ever. Even the ground temperature stays above that. I'm wondering whether this might play a role in enabling this oven to work so well.

All the best

Stephen

Thank you for the update and well done to get your concept up and running.