I've been converting a 150,000 BTU natural gas salamander heater to propane.
I finally got everything together Friday, and turned it on. Basically I
have produced an afterburner. Way too much gas flow -- sounds like an F-16.
BIG flame and it cycles on the overheat safety, not the thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes. I
machined a new one with #50 holes, based on the cross sectional area of the
holes and the relative heat content of natural gas vs. propane (because I
left the NG regulator inside, the pressure supplying this jet block is the
same as it was for NG). I figured that this would be a good first attempt.

Clearly not even close to the right size. I can deal with that, but there
is an interesting conundrum here. In trying to tame the flame, I throttled
it down with the propane tank valve, and got down to a more reasonable flame
size. For safety, I've put a CO detector in the garage during these
experiments, and with the BIG flame, it detects no CO. However, with the
throttled flame, the CO level rises fairly rapidly, and there is obviously
something in the air that strongly irritates the nose and throat.

I would have thought that the opposite would happen. The air flow is fixed
by the electric blower, so with too much gas there should be incomplete
combustion. There is quite a bit of yellow tips on the flames with the full
gas flow, which seems consistent with this hypothesis. However, with less
gas, there should be more complete combustion. There are fewer yellow
flames when I throttle back the gas flow, but I begin to get a lot of CO,
and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back, there
is very little yellow, the heater cycles on the thermostat, not the safety
cutout, and all seems fine except that it's trying to kill me. Thoughts
from you experts?

--
Bob (Chief Pilot, White Knuckle Airways)

"Bob Chilcoat" wrote in message
...
I've been converting a 150,000 BTU natural gas salamander heater to
propane. I finally got everything together Friday, and turned it on.
Basically I have produced an afterburner. Way too much gas flow -- sounds
like an F-16. BIG flame and it cycles on the overheat safety, not the
thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes. I
machined a new one with #50 holes, based on the cross sectional area of
the holes and the relative heat content of natural gas vs. propane
(because I left the NG regulator inside, the pressure supplying this jet
block is the same as it was for NG). I figured that this would be a good
first attempt.

Clearly not even close to the right size. I can deal with that, but there
is an interesting conundrum here. In trying to tame the flame, I
throttled it down with the propane tank valve, and got down to a more
reasonable flame size. For safety, I've put a CO detector in the garage
during these experiments, and with the BIG flame, it detects no CO.
However, with the throttled flame, the CO level rises fairly rapidly, and
there is obviously something in the air that strongly irritates the nose
and throat.

I would have thought that the opposite would happen. The air flow is
fixed by the electric blower, so with too much gas there should be
incomplete combustion. There is quite a bit of yellow tips on the flames
with the full gas flow, which seems consistent with this hypothesis.
However, with less gas, there should be more complete combustion. There
are fewer yellow flames when I throttle back the gas flow, but I begin to
get a lot of CO, and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back, there
is very little yellow, the heater cycles on the thermostat, not the safety
cutout, and all seems fine except that it's trying to kill me. Thoughts
from you experts?

I wouldn't assume the NG regulator is OK for propane, given the difference
in inlet pressure.

Bob Chilcoat wrote:
I've been converting a 150,000 BTU natural gas salamander heater to propane.
I finally got everything together Friday, and turned it on. Basically I
have produced an afterburner. Way too much gas flow -- sounds like an F-16.
BIG flame and it cycles on the overheat safety, not the thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes. I
machined a new one with #50 holes, based on the cross sectional area of the
holes and the relative heat content of natural gas vs. propane (because I
left the NG regulator inside, the pressure supplying this jet block is the
same as it was for NG). I figured that this would be a good first attempt.

Clearly not even close to the right size. I can deal with that, but there
is an interesting conundrum here. In trying to tame the flame, I throttled
it down with the propane tank valve, and got down to a more reasonable flame
size. For safety, I've put a CO detector in the garage during these
experiments, and with the BIG flame, it detects no CO. However, with the
throttled flame, the CO level rises fairly rapidly, and there is obviously
something in the air that strongly irritates the nose and throat.

I would have thought that the opposite would happen. The air flow is fixed
by the electric blower, so with too much gas there should be incomplete
combustion. There is quite a bit of yellow tips on the flames with the full
gas flow, which seems consistent with this hypothesis. However, with less
gas, there should be more complete combustion. There are fewer yellow
flames when I throttle back the gas flow, but I begin to get a lot of CO,
and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back, there
is very little yellow, the heater cycles on the thermostat, not the safety
cutout, and all seems fine except that it's trying to kill me. Thoughts
from you experts?

Sounds like fun! Good first approximation approach on the orifice size,
except that for small nozzles the flow is not linear with area. It
involves voodoo and math that made my head hurt.

Regarding the incomplete combustion, my WAG is that with low pressure
the gas mixture is too lean to keep all of itself at combustion
temperature. You might try throttling the air intake; although a waste
gate would be better it is a lot of fooling around for a trial.

--
Fred R
"It doesn't really take all kinds; there just *are* all kinds".
Drop TROU to email.

The ones I have seen that are propane look and sound like an afterburner
Stock. Big loud swirling blue flame. Maybe you do have it right. Loud
enough that it is hard to talk around if not impossible. But they make
heat really fast so it only has to be on for a minute or two.

I think I'd call the guys at
http://www.portableheaterparts.com/contact.html

They have been quite helpfull to me in the past. Give them the specs on
your unit, they may have the right orifice on the shelf. Prices have
been fair. I'd be more comfortable with knowing I had the RIGHT orifice,
and that I wasn't overdriving the heater.

Bob Chilcoat wrote:

I've been converting a 150,000 BTU natural gas salamander heater to propane.
I finally got everything together Friday, and turned it on. Basically I
have produced an afterburner. Way too much gas flow -- sounds like an F-16.
BIG flame and it cycles on the overheat safety, not the thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes. I
machined a new one with #50 holes, based on the cross sectional area of the
holes and the relative heat content of natural gas vs. propane (because I
left the NG regulator inside, the pressure supplying this jet block is the
same as it was for NG). I figured that this would be a good first attempt.

Clearly not even close to the right size. I can deal with that, but there
is an interesting conundrum here. In trying to tame the flame, I throttled
it down with the propane tank valve, and got down to a more reasonable flame
size. For safety, I've put a CO detector in the garage during these
experiments, and with the BIG flame, it detects no CO. However, with the
throttled flame, the CO level rises fairly rapidly, and there is obviously
something in the air that strongly irritates the nose and throat.

I would have thought that the opposite would happen. The air flow is fixed
by the electric blower, so with too much gas there should be incomplete
combustion. There is quite a bit of yellow tips on the flames with the full
gas flow, which seems consistent with this hypothesis. However, with less
gas, there should be more complete combustion. There are fewer yellow
flames when I throttle back the gas flow, but I begin to get a lot of CO,
and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back, there
is very little yellow, the heater cycles on the thermostat, not the safety
cutout, and all seems fine except that it's trying to kill me. Thoughts
from you experts?

On Sun, 19 Mar 2006 07:51:25 -0500, "Bob Chilcoat"
wrote:

I've been converting a 150,000 BTU natural gas salamander heater to propane.
I finally got everything together Friday, and turned it on. Basically I
have produced an afterburner. Way too much gas flow -- sounds like an F-16.
BIG flame and it cycles on the overheat safety, not the thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes. I
machined a new one with #50 holes, based on the cross sectional area of the
holes and the relative heat content of natural gas vs. propane (because I
left the NG regulator inside, the pressure supplying this jet block is the
same as it was for NG). I figured that this would be a good first attempt.

Clearly not even close to the right size. I can deal with that, but there
is an interesting conundrum here. In trying to tame the flame, I throttled
it down with the propane tank valve, and got down to a more reasonable flame
size. For safety, I've put a CO detector in the garage during these
experiments, and with the BIG flame, it detects no CO. However, with the
throttled flame, the CO level rises fairly rapidly, and there is obviously
something in the air that strongly irritates the nose and throat.

I would have thought that the opposite would happen. The air flow is fixed
by the electric blower, so with too much gas there should be incomplete
combustion. There is quite a bit of yellow tips on the flames with the full
gas flow, which seems consistent with this hypothesis. However, with less
gas, there should be more complete combustion. There are fewer yellow
flames when I throttle back the gas flow, but I begin to get a lot of CO,
and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back, there
is very little yellow, the heater cycles on the thermostat, not the safety
cutout, and all seems fine except that it's trying to kill me. Thoughts
from you experts?

Some gas sensors are not very specific, so your CO sensor may be
sensing something else. For example, this sensor
http://www.microchemical.com/co.htm
senses CO *and* NO2.

Usually, CO is associated with incomplete combustion, or "yellow
flame". Blue flames run at higher temperature and produce more
NOx. NOx is an irritant. NO2 combines with moisture to form nitric
acid. If the moisture is found in your nose and throat, that could
indeed be irritating!

On Sun, 19 Mar 2006 07:51:25 -0500, "Bob Chilcoat"
wrote:

I've been converting a 150,000 BTU natural gas salamander heater to propane.
I did the same thing with some naturally aspirated NG heaters. Take
out the orifice and carry it to a propane supplier. Tell them what the
heat output rating for the device is (mine were 25K and 30K). Have him
look up the proper orifice size for that heat output with LP fuel gas
and drill a replacement orifice to the appropriate size. Worked great
for me.
What you have now is basically a weed burner in a salamander body. I
have a 30K-55K LP heater that was purchased in that configuration. I
also have the 500K weed burner. You can tell when the heater is on
since it has a fan, but it's nowhere near as loud as the weed burner
at full bottle pressure.

The NG regulator is a final reduction with a 1/2" inlet pressure and a lower
outlet pressure (haven't measured it). Since the pressure from the propane
regulator at the tank is 11", and 1/2 psi is 14", the NG regulator should
just drop the propane pressure to whatever the normal NG pressure is to the
orifice block. So I need to figure out the equivalent orifice size for LP
at the same pressure as the NG. If I took out the NG regulator, I could
just use the proper LP jet (which is supplied at 11"), which I can get from
the manufacturer, or machine one with the same size holes. I'm trying to
save the hassle of putting in a length of tubing where the NG regulator is
(between the orifice supply tube and the solenoid valve). OTOH, it might be
a lot easier to do that in the long run, but perhaps not as interesting.
Drilling ten radial holes in my shop-built orifice block involved using the
indexing head that I've only used once before. Kinda fun (HSM content).

--
Bob (Chief Pilot, White Knuckle Airways)


"ATP*" wrote in message
...

"Bob Chilcoat" wrote in message
...
I've been converting a 150,000 BTU natural gas salamander heater to
propane. I finally got everything together Friday, and turned it on.
Basically I have produced an afterburner. Way too much gas flow --
sounds like an F-16. BIG flame and it cycles on the overheat safety, not
the thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes. I
machined a new one with #50 holes, based on the cross sectional area of
the holes and the relative heat content of natural gas vs. propane
(because I left the NG regulator inside, the pressure supplying this jet
block is the same as it was for NG). I figured that this would be a good
first attempt.

Clearly not even close to the right size. I can deal with that, but
there is an interesting conundrum here. In trying to tame the flame, I
throttled it down with the propane tank valve, and got down to a more
reasonable flame size. For safety, I've put a CO detector in the garage
during these experiments, and with the BIG flame, it detects no CO.
However, with the throttled flame, the CO level rises fairly rapidly, and
there is obviously something in the air that strongly irritates the nose
and throat.

I would have thought that the opposite would happen. The air flow is
fixed by the electric blower, so with too much gas there should be
incomplete combustion. There is quite a bit of yellow tips on the flames
with the full gas flow, which seems consistent with this hypothesis.
However, with less gas, there should be more complete combustion. There
are fewer yellow flames when I throttle back the gas flow, but I begin to
get a lot of CO, and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back,
there is very little yellow, the heater cycles on the thermostat, not the
safety cutout, and all seems fine except that it's trying to kill me.
Thoughts from you experts?

I wouldn't assume the NG regulator is OK for propane, given the difference
in inlet pressure.

Don,

I think you've identified why throttling back the gas pressure produces that
irritating contaminant. I hadn't thought about NO2 (although is should
have, I've been involved with N2O for years in my anesthesia work -- related
oxide of nitrogen). It makes perfect sense that with less gas and a fixed
airflow, the temperature should rise. I guess there's an optimal
temperature range that minimizes CO and NO2 production. Now I have to
figure out what that is. Thanks.

--
Bob (Chief Pilot, White Knuckle Airways)


"Don Foreman" wrote in message
...
On Sun, 19 Mar 2006 07:51:25 -0500, "Bob Chilcoat"
wrote:

Some gas sensors are not very specific, so your CO sensor may be
sensing something else. For example, this sensor
http://www.microchemical.com/co.htm
senses CO *and* NO2.

Usually, CO is associated with incomplete combustion, or "yellow
flame". Blue flames run at higher temperature and produce more
NOx. NOx is an irritant. NO2 combines with moisture to form nitric
acid. If the moisture is found in your nose and throat, that could
indeed be irritating!

"Bob Chilcoat" wrote in message
...
The NG regulator is a final reduction with a 1/2" inlet pressure and a
lower outlet pressure (haven't measured it). Since the pressure from the
propane regulator at the tank is 11", and 1/2 psi is 14", the NG regulator
should just drop the propane pressure to whatever the normal NG pressure
is to the orifice block. So I need to figure out the equivalent orifice
size for LP at the same pressure as the NG. If I took out the NG
regulator, I could just use the proper LP jet (which is supplied at 11"),
which I can get from the manufacturer, or machine one with the same size
holes. I'm trying to save the hassle of putting in a length of tubing
where the NG regulator is (between the orifice supply tube and the
solenoid valve). OTOH, it might be a lot easier to do that in the long
run, but perhaps not as interesting. Drilling ten radial holes in my
shop-built orifice block involved using the indexing head that I've only
used once before. Kinda fun (HSM content).

--
Bob (Chief Pilot, White Knuckle Airways)

Maybe there's a reason why the manufacturers have it set up without the NG
regulator. I think there are a lot of potentially complicated combustion
engineering issues here and the stakes are very high.


"ATP*" wrote in message
...

"Bob Chilcoat" wrote in message
...
I've been converting a 150,000 BTU natural gas salamander heater to
propane. I finally got everything together Friday, and turned it on.
Basically I have produced an afterburner. Way too much gas flow --
sounds like an F-16. BIG flame and it cycles on the overheat safety, not
the thermostat.

OK, so I didn't calculate the resized orifices correctly. The jet block
consists of a cylindrical brass piece with ten radial #40 drill holes.
I machined a new one with #50 holes, based on the cross sectional area
of the holes and the relative heat content of natural gas vs. propane
(because I left the NG regulator inside, the pressure supplying this jet
block is the same as it was for NG). I figured that this would be a
good first attempt.

Clearly not even close to the right size. I can deal with that, but
there is an interesting conundrum here. In trying to tame the flame, I
throttled it down with the propane tank valve, and got down to a more
reasonable flame size. For safety, I've put a CO detector in the garage
during these experiments, and with the BIG flame, it detects no CO.
However, with the throttled flame, the CO level rises fairly rapidly,
and there is obviously something in the air that strongly irritates the
nose and throat.

I would have thought that the opposite would happen. The air flow is
fixed by the electric blower, so with too much gas there should be
incomplete combustion. There is quite a bit of yellow tips on the
flames with the full gas flow, which seems consistent with this
hypothesis. However, with less gas, there should be more complete
combustion. There are fewer yellow flames when I throttle back the gas
flow, but I begin to get a lot of CO, and that irritating byproduct.

Is my logic wrong here? I'm confused. With the gas throttled back,
there is very little yellow, the heater cycles on the thermostat, not
the safety cutout, and all seems fine except that it's trying to kill
me. Thoughts from you experts?

I wouldn't assume the NG regulator is OK for propane, given the
difference in inlet pressure.

On Mon, 20 Mar 2006 09:49:21 -0500, "Bob Chilcoat"
wrote:

Don,

I think you've identified why throttling back the gas pressure produces that
irritating contaminant. I hadn't thought about NO2 (although is should
have, I've been involved with N2O for years in my anesthesia work -- related
oxide of nitrogen). It makes perfect sense that with less gas and a fixed
airflow, the temperature should rise. I guess there's an optimal
temperature range that minimizes CO and NO2 production. Now I have to
figure out what that is. Thanks.

Right, Bob. Combustion products NOx and CO are like radio buttons:
reduce one, raise the other. Some industrial burner controls use
oxygen sensors in the flue gas to control fuel-air mix to meet
standards for NOx and CO emissions. I think auto engines do that as
well, now that they have engine control computers. O2 content
of flue gas seems to be a useful indicator of combustion condx. It's
an indirect and inferential measurement, but it seems to work. Oxy
sensors are cheap, and they're robust in a fluegas environment.

I'm no expert in the field but you should be able to find data from
ASHRAE. If the process was oil combustion I am sure you could get
data from Dr. Tom Butcher at Brookhaven Nat'l Labs. You might try
pinging Dr. Ulrich Bonne at Honeywell Labs in Plymouth, MN. Ulrich
probably knows as much about gas combustion as anyone on the planet.


I don't know if he'd respond or not, but he's a good guy so he might.
Tell him Dr. Foreman referred you to him. That might just evoke a
very informative response because he knows damned well I'm not a
PhD and he purely hated it when I was assigned to conduct a critical
tech review of his work once upon a time..... G