A Research Organization
The liquid-vapor separator has three towers on it. The first (left when facing front of machine) is 8 inches high and has the pressure and liquid level sensors on it. The second is 13 inches high, and is the "flame modifier" tank. The third is also 13 inches high and is designated as the "back arresting" solution container (backfire container, or bubbler in my electrolyzer design).
The two end towers (left and right) are further connected at their bottoms (just above the liquid-vapor separator) by a second transverse container 4 inches in diameter and 17 inches long. The actual connection is by two short 4 inch tubes. I'd call this tube a "surge" tube, I think it is meant to allow storage space for pressurized gas and/or to prevent serious over-filling of water from affecting the gas quality too much by allowing water/electrolyte to be carried over into the torch hose.
Personally, I see the surge tube as way too large, or un-needed altogether. In my opinion the whole arrangement could have been simplified and/or modified to be more versatile and effective. For example, it would be nice to be able to switch between pure and modified gas with a valve (without having to drain the modifier tank to eliminate the modifier from the flame); or even to be able to run pure and modified gas at the same time (with two torches). Also, I see no reason that the water fill for the electrolyzer couldn't have been mounted on the sensor tower, leaving the back arresting tower simpler and cutting out the surge tube entirely. But I will qualify my opinions by saying I don't know exactly all the plumbing arrangements inside these various tubes. My electrolyzer designs are simpler, more versatile and fully documented so that people aren't guessing about their function.
The gas seems to go up the two hoses on the top of the cells (connecting all the cells together), through the liquid-vapor separator, through the modifier tank first, then go back down (with some sort of inner U-tube) and up through the back arresting tank, then out the hose to the torch.
Water for the electrolyzer cells is put in the back arresting tower, where the water is poured through a down tube directly to the liquid-vapor separator; bypassing the back arresting apparatus in the back arresting tower.
The water level probes in the sensor tower indicate that the maximum liquid level in the liquid-vapor separator to be about mid level, lowest about 1 inch off the bottom.
Water for the back arresting tank is supposed to be filled by disconnecting the gas-out hose (found at the top-backside of the tank) and using a small flexible hose to pour water into the tank. Top cover of the machine must be removed to do this. We were unable to get our hose past the fitting; so we drilled and tapped an 1/8 inch plug in the top of the back arresting tank. We then filled the back arresting tank till water came out the drain (was about 1/2 quart), then closed the drain and added 1/2 quart more (0.5 liter) that the manual specified. It was unclear whether this second 0.5 liter was needed, so we added it anyway.
Advise for operators of BN 1000E operators: drill new filler plug to the left of main water filler facing from front of machine), so water being poured into the plug hole (for the back arresting tank liquid level) doesn't drain into the gas-out hose (water in your torch hoses is a bother, causes your flame to spit and go out). And always drain your back arresting tank before re-filling it, or it will be impossible for you to know what the liquid level is (need only 0.5 liter above filler).
I think it'd be a good idea to drill and tap a "sight tube" (with a floating ball) on the side of the "back arresting" tower; then you could see at a glance your back-arresting water level. Don't drill the lower hole too low or you'll be below the bottom plate and cause the gas to by-pass the backfire arrester. You want your liquid level to be well above the diffuser plate; you'll see the diffuser plate (has holes in it) when you drill the new hole in the top of the gas-out tower. But you don't want your liquid level to be too high or you will get water splashing up and out into your torch hose.
Although the BN 1000E has proven that it can take a backfire without blowing up (we accidentally backfired it by turning the torch off too slow and we discovered that we had no water in the (separately filled) back arresting tank) I notice the fittings in the plastic hoses are now seeping a bit of fluid (even though this machine has not reached pressures over 8 psi); this seepage could have been caused by the backfires that this machine has had it the past, with no water in the back arrester. My recommendation is to keep water in the back arresting container to prevent the electrolyzer and hoses from having to deal with the shock of a backfire.
Note: My design does not have the dozens of fittings that this electrolyzer has. Every fitting is a potential leak of fluid or gas.
The modifier tank is to be filled with water first, to above the drain level, then drained, plugged and modifier added (no more than about 1/2 liter or 1/2 quart). To be sure that all of the modifier has been removed, it may be necessary to add water after draining the modifier, let sit for at least 1/2 hour and then drain it again. Most flammable fluids float on water.
Different modifiers will give different flame characteristics, and I discovered it may be necessary to add oxygen to get the best performance out of the modifiers. Some examples of modifiers include; gasoline (will change it's characteristics as the "light ends" evaporate first), diesel, acetone, benzene, white gas, all types of alcohol, and many other flammable liquids. You could just use water in this tank too, and it'd work as a second backfire arrester.
A temperature sensor (resistive?) on the fourth electrolysis cell from the front. Disconnecting this sensor causes the temperature readout to go to "00" and doesn't immediately shut down the machine. The temperature seems to be shutting off at about 35°C. The temperature readout is in °C even though the button you press to check the temperature is labeled "F".
A pressure sensor (transducer?) on the third tower, (back left of machine). Disconnecting this sensor doesn't immediately shut down the machine, the "actual" pressure reading goes to "00". When this sensor is disconnected the pressure continues to rise past the computer set maximum. The pressure rises VERY fast and may rise enough to burst the BN 1000E even with the torch running.
A liquid level sensor, which is three probes put down into the solution. They seem to be simple copper rods, which form some sort of on/off switch. Examination of the LED's on the front panel indicate that they are arranged in three groups. So I expect that as each probe (three different lengths) touches the solution, the appropriate group of LED's light up. I note that the liquid level is supposed to be about even with the hoses coming in from the cells; highest level about 2.5 inches off the bottom of the liquid-vapor separator, lowest level about 3/4 inch off the bottom of the liquid-vapor separator. Disconnecting this sensor does not immediately shut off the machine, the panel of LED's does go out.
Note: That in our testing of this BN 1000E, we found one LED to be "blown" and the three probes of the liquid level probe to all be shorted to each other; thus I assume the liquid level probes to not be working as originally designed or intended. At the moment, when the liquid level goes to below the bottom probe, all but two of the LED's go out, and the "low water" light comes on (which is good).
Note: We operated the machine with the liquid level 2 inches higher than the highest set point with absolutely no problem.
Operation notes: The Brown's Gas 1000E makes significant noise during operation, mostly from the fan; it is so loud that we can't even hear the noise of the transformer. The problem is noise pollution, making normal conversation difficult and making stress.