Batch box modification to view flame.

I am planning to build a rocket mass heater within the next year.  I have been thinking of a design that puts the burn chamber directly below the heat riser to allow for openings to see the flame and get some fast radiant heat with the fire lit.
I made a crude experiment with some galvanized duct and an empty refrigerant bottle.  After observation of the rocket outdoors, i added a bell with left over 9x14 duct.  Its really working well so far.  
I'm thinking i should do a version out of mud outside to really test it, then do the real deal out of masonry.  I'm wondering if this has been done before and what issues might arise?
I'd like to see a lot of the flame in the final build, maybe openings on 3 sides?

That seems like a worthy goal. I am curious, have you read through Ianto Evans, erika and Ernie, or Peter?  Please be wary of getting advice from YoTubers,etc.  Advice from any of the aforementioned experts would be ideal

Hi Ben;
Nice little rocket-style stove you built there.
Building rockets with glass has been tried before.  They are beautiful while burning, and as a rocket stove, they work great.
Other than the high costs of fire glass (average $10 a square inch), and the constant danger of the glass getting broken while a fire is burning.
As a mass heater, ultimately, you will not be happy.
The problem is that a lot of heat is escaping through the glass.
Here's what will happen if you try to turn that into a mass heater.
To start, as soon as you divert the exhaust from leaving straight up from the riser, your glass will blacken, and you will still see flames, but not the way you're hoping.
Next, because heat is escaping the core and not all of it is going through the riser, you will not reach the high temperatures required for a full burn.
You'll create ash build-up, and the infamous creosote will form, creating a chimney fire in the making.

If you want glass, consider building a proven Shorty Core; she can have a nice, large window to enjoy the fire.
Also, there are the double shoebox designs (DSR) that offer a spectacular view of the double ram's horn fire in the upper box.
Rocket scientist Fox James has some great videos of these designs.

Any batch box of standard design can have glass in the door to see the flame. You can make a radically different design if you want, but there is no assurance that it will actually burn clean and efficient without professional testing equipment and probably many tweaks to fine-tune it.

One of Peter van den Berg's lines of research was the "double shoebox rocket" (DSR) which used a port in the ceiling of the firebox, so looking at those would help you understand the issues. Fox James on Permies has done at least one experiment with a ceiling port and extreme flame visibility. I think there were other experimenters on Donkey's forum on Proboards who worked with ceiling ports.

The main issue I can see from your picture and description, is the amount of unburnt fuel exiting the riser?

I have developed a small vortex stove, it is built from insulating material but uses large viewing windows and a large ceramic glass hot plate to disperse the heat into a space. (Space heater)
The placement of the glass allows huge amounts of heat to be released which in turn heats a space very efficiently while keeping the chimney temperatures very low, resulting in a super efficient hot plate and space heating stove but, nor a mass heater as such.

So drawing the flame path away from the glass (like most stoves do) might work but if the flame is going straight up without
any turbulent hot, gas, air mixing, then it wont be very efficient at all?

However we are here to help, so if you want to offer more details then perhaps we can give you more food for thought…..

Thanks for the feedback!  I'm going to look thru Fox's channel tonight to better understand some of this.  I do think i see what you mean about the need for turbulance.  My goal is to build before next winter, so I'll be studying, experimenting, and reporting back as I go.

The picture was at low light, the flame maybe looks bit bigger than it was.  I really expected the Square duct bell to get much hotter, based on how much flame came out of the riser before i added it.  Not sure if thats good or bad?   I have a cheap thermal imager i've been playing with. Soon as i figure out how to get the pics off, ill share them.

Turbulence is one of the „3 Ts“ for complete combustion that I consider the fundamentals for anything calling itself a rocket.
Time, temperature and turbulence that need to be high enough to ensure that all fuel (including smoke and creosote) is burnt before exhausting.
It’s fun to experiment. But it’s key to built on top what works and not repeat what has proven to not work.