[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Heat gradients
>From: "James Purchase" <jpurch at interlog_com>
>Date: Wed, 6 Jan 1999 12:21:11 -0500
>Mr. Booth, please get your thermometer out and let us know what kind of
>heating gradient is being produced in the substrates of your tanks by the
Mr. Booth?? Well, my Dad doesn't know jack about heating cables but perhaps
I, humble George, can be of some assistance.
I was surprised to find a wealth of information about heating coils in The
Here is one of many archived articles that I found, reproduced for
your convenience. I've also added a bit at the end to bring closure
to the information.
Heating coil temperatures
by George Booth
Date: 31 May 1994
We made some measurements on some substrate heating coil setups
recently and thought the net may be interested.
We used a heat sensing IC, an LM344. This device gives an output
voltage equal to the temperature in degrees Kelvin: 0 degrees C equals
273 degrees K. Room temperature, 77 F or 25 C, is 298 K. There is
another device, an LM34, which is proportional to degrees C, but I
couldn't find one locally. Digikey and other electronic suppliers
The LM34 is better since you will have better resolution with a cheap
digital voltmeter; I had to use an expensive 4 1/2 digit HP voltmeter
to make the following measurements. With the LM344, room temp is 2.98
volts and you need a precise voltmeter to resolve tenths of a degree
(2.981, 2.983, etc). With the LM34, I conjecture room temp is a
reading of 2.50 volts and a cheap voltmeter can easily resolve tenths
of a degree.
Thanks to Lawrence Smith (lawrence-at-msc.cornell.edu) for giving me the
details on these devices.
We recently renovated our 100 gallon discus tank and added heating
coils to it. We did the Dan Resler thing; used 100 watt Dupla coils
run by a commercial transformer and controlled by a light timer.
We first set the coils up with a 18v 4a transformer instead of 24v
since that's what I had laying around. This gave us just over 50
watts of heat. The 100 watt Dupla coil is about 22 feet long and gave
us 4 transversals of the tank with the coils spaced 3 1/2 inches
apart. For reference, the Dupla coil anchors dictate a spacing of
The timer was set to turn on and off every hour (1 hour on, 1 off).
This caused the water temperature to rise just a tad over the setting
of the Ebo-Jaeger heaters, creating the situation of having the
substrate coils supplying most of the heat to the water and the
Ebo-Jaeger heaters being used for backup.
We measured the temperature in the 3" deep substrate after the coils
were on for 1 hour. This is what we found:
Water temperature : 29.0 C
Top of gravel, 2 1/2" from cable : 29.2
Middle of gravel, 1" from cable : 29.5
At the cable : 30.1
At the bottom, 1/2" below cable : 29.9
At the bottom, between cables : 28.6
With 50 watts of heat and 3 1/2" between cables, there is a 0.9 degree
C temperature difference from the top of the gravel to the cable and a
1.5 degree temperature gradient between cables, giving rise to
hot/cold zones. I suppose this will create some convection currents
but it's hard to measure.
Once the 24v 5a transformer arrived, we remeasured the temperatures.
With the higher heat, we had the coils on for 1 hour and off for 2
hours. This seemed to maintain the temperature as before. With a 50%
duty cycle, we noticed the water temperature was getting to 29.5 C.
We need to set the Ebo-Jaegers at a lower temp and watch what the
coils will do on their own.
Anyway, with the 100w coils on for 1 hour, this is what we found:
Water temperature : 29.2 C
Top of gravel, 2 1/2" from cable : 29.4
Middle of gravel, 1" from cable : 30.1
At the cable : 31.5
At the bottom, 1/2" below cable : 30.8
At the bottom, between cables : 28.5
With 100 watts of heat and 3 1/2" between cables, there is a 2.1
degree C temperature difference from the top of the gravel to the
cable and a 3.0 degree temperature gradient between cables, giving
rise to hot/cold zones. I suppose this will create some convection
currents but it's hard to measure.
The 90 gallon tank we set up over 2 1/2 years ago has a complete 250
watt Dupla cable setup with a Dupla transformer and controller. The
250 watt coil is about 60 feet long and gave us 14 transversals of the
tank with the coils spaced 1 1/4" apart. For reference, the Dupla
coil anchors dictate a spacing of 1 3/4". We felt the extra wattage
was needed due to a glass tank with an open top with a water temp of
82 C in a cooler room. We didn't realize the cable would be so long.
It still requires an extra 250 watt Ebo-Jaeger heater to maintain the
temperature on cold winter days.
With the 250w coils on for 1 hour, this is what we found:
Water temperature : 28.0 C
Top of gravel, 3" from cable : 28.0
Top 1/3 of gravel, 2" from cable : 30.6
Bot. 1/3 of gravel, 1" from cable : 31.4
Bottom, at the cable : 33.3
Bottom, between cables : 33.2
With 250 watts of heat and 1 1/4" between cables, there is a 5.3
degree C temperature difference from top to bottom, but no temperature
gradient between cables (no hot/cold zones). I suppose this will
create some convection currents but it's hard to measure.
The lower wattage setup is based on Dennerle recommendations and the
higher wattage setup is based on Dupla (sort of :-). We know the high
substrate heat setup produces excellent results (long term stability).
We will have to wait for awhile to assess the lower wattage results.
Back to the present (1/6/99). The "lower wattage" results were NOT very
impressive. We had algae and growth problems in the lower wattage tanks
after about 1 year, just as we had in the "Almost Optimum Aquarium" a few
years before (plain gravel and laterite, NO heating coils).
We had to tear down all the aquariums in May of 1996 for home remodeling and
redid them all with "high wattage" cables. The two 100 gallon tanks got 150w
coils (Rainbowfish tank kept at 76F) and 200w coils (Discus tank kept at 82F).
We scrapped the 90 gallon SST and picked the best equipment to set up the
120 gallon Rainbowfish tank. It received the 250w coils and is kept at 76F.
All the tanks have been doing great since they were redone. I would recommend
the higher wattage scheme over the lower wattage scheme based on our non-
George Booth, Ft. Collins, Colorado (booth at frii_com)
Back on-line! New URL! Slightly new look! Same good data!