[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

DIY info needed

Hi Martyn,

> I have found many bits and pieces, however ,those of us who are not
> well versed in electrical engineering could benefit from a
> consolidated and detailed description of the parts required, pertinent
> specs as well as a schematic description.

Here are two articles from my archives -- I think they are in the Krib
archives somewhere but don't know for sure. The first was written by a
CS prof with zero electrical experience.  He built a system for
himself, so the instructions are pretty good. He used Dupla cables and
did the other stuff himself.  The second was written by a EE in
Germany and describes making the cables yourself.  A bit more
technical but not too hard to understand.  

Also check out my website for other heating cable info.

George Booth


From: dresler at cabell_vcu.edu (Dan Resler)
Date: Mon, 13 Dec 1993 22:57:16 GMT
Subject: a heating cable system

I recently set up a hi-tech 90gal plant tank following the Dupla & Booth
plans. Since there's been several queries concerning heating cable systems I
decided to report on the setting up and installing of my cable system.

The Cables

I tend to err on the safe side when it comes to anything I put into the
water, so I decided to pay the extra money and get the Dupla cables (as
compared to using standard wire as recently proposed by Uwe Behle). I'm
using the 24 volt Dupla 300 cables which Dupla says are 7m long and are rated
at 100 watts. A resistance measurement gives them an actual rating of about
94 watts, while the actual length of the `hot' section is 6.96m (22'10").

My objective was to get the minimum wattage that would provide water flow
through the substrate; I did not want the cables to contribute much to
heating the tank as I anticipate trouble with keeping the tank cool in the
summer. Unfortunately, as has been mentioned in this group previously, the
lower the power rating of the cables the shorter Dupla makes them (this
applies to both the `hot' and `cold' parts of the cables; my leads to the
transformer box were about 3' too short). IMHO, a cable length of ~7m is the
absolute minimum for a tank with a 4' X 18" footprint.

Cables 7 meters long allow for 6 traversals of the tank bottom with ~6cm
between the cables. The Dupla cable anchors are ridiculously expensive, so I
devised my own using PVC pipe and electrical cable ties. I cut 1/2" PVC
pipe (use a bandsaw!) in half lengthwise to provide 6 ~17" supports across
the width (i.e. front to back) of the tank. Then two small holes were drilled
on each side of where a cable would cross the supports, and cable ties were
pushed through these holes and around the heating cables to secure them to the
half-pipe sections. Since snugging the ties prevented the cables from moving, I
was able to construct everything outside of the tank (a big advantage with a
24" high tank) and then gently lift the whole system in.

The Power Supply

The Dupla cable controller was way outside my budget; I decided early on
that I would have to build my own. This posed a problem, however, as I know
very little about electronics! After many discussions with net people and
attendance at the Booth School for the Electrically Impaired (with guest
lectures by Uwe Behle and Anthony Tse), I decided to build a very simple
transformer box with no temperature controller; here's a rough ASCII schematic
of what I built:
                               *      *
           O----+--------------+      +----------fuse------O
                |              *      *
  to cables     |        24v   *      *  120v              to mains
              lamp   secondary *      * primary
                |              *      *
                |              *      +--------------------O
           O----+--------------+      *
                               *      *         earth------O

My first design involved adding a light dimmer on the primary side to
control the power delivered to the cables. I experimented with several
light and ceiling fan dimmers but they all proved unsatisfactory; the EE
types on the net I consulted with reckoned they were throwing out funny
wave forms that messed with the transformer (I was unable to confirm this
since I don't have access to a 'scope).

If you build anything like this, be sure to add a slo-blo fuse to the hot
(black) lead to the primary. I'm told transformers do fail. (BTW, please do
not attempt to build something like this unless you're sure of your design
and you know what you're doing; I didn't, so I spent lots of bandwidth asking
questions and getting people to check my very simple design.)

The Controller

Since I didn't know how to build a temperature controller and I couldn't
afford to buy one, I decided to use a simple mechanical timer to control
the cables. After much searching I found one that allowed up to 24 on-off
cycles in 24 hours (the timer must be able to handle an inductive load; all
of the ones I've seen labeled `heavy-duty' can handle such loads).

The Results

I installed the cables under about 3.5" of gravel. With the cables off, the
water measured 72 F. I first set the timer for 12 on-off cycles in 24 hours
(i.e. 1 hour on followed by 1 hour off). Within 10 hours the tank stabilized
at 76 F (with lights off the entire time). With the cables on (45 minutes into
their 1 hour cycle) the temperature at the gravel surface was measured to be
the same as the water temperature near the top. I took 5 measurements around
the tank about 1" deep into the gravel and got a reading of 76 F (a couple
might have been closer to 76.5 F). To confirm that the cables were indeed
working I turned them off at 10pm one evening; by 7am the water temperature
had fallen to 72 F.

I wished to maintain the water temperature at about 79 F, so I added 2 75
watt Ebo-Jager submersible heaters and began experimenting with the heater
thermostats and the cable on/off cycles. In order to provide a temp drop of
1 - 2 degrees (I might go to 2-3 degrees) at night I turned the cables off
from 2 - 5 am and set heater A to go on at ~76 F (in case of a cold spell)
and heater B at ~78 F. So far only heater B has been on in the morning and the
tank has stabilized at 77.5 to 78 F at 8am and 79 F at 9pm. The cables are
on 75% of the time: from 5am - 8pm, 9 - 10pm, 11 - 12pm, 1 - 2am. The ambient
room temperature varies from 70 - 72 F, I am using a trickle filter and an
Ehiem canister, 160w fluorescent lighting, and the tank is partially open at
the top with no insulation at the bottom.

The Cost

Dupla 300 cables              US$95.00
PVC pipe & cable ties             4.00
Transformer box                 ~50.00
24 hour timer                     9.00

The most difficult parts to find were the 24v, 4amp transformer (I had to
order it) and the timer (most I saw had only 2 - 4 on/off cycles). I
eventually stumbled upon an `Ingraham Model 12-201E Multiple Program Timer'
from Toastmaster, Inc, at Wal-Mart!

If one was willing to forgo the Dupla cables, it then becomes possible to
have a working heating cable system for under $75.

Dan Resler                             
Dept. of Mathematical Sciences        email: dresler at cabell_vcu.edu
Virginia Commonwealth University
Richmond, Virginia, USA 23284-2014

From: uweb at hpbidrd1_bbn.hp.com (Uwe Behle)
Date: Mon, 18 Oct 1993 08:38:06 GMT
Subject: Re: Re: [F][Plants] Heating coil effects - discussion

There seems to be a lot of misconceptions about the safety of home-built
heating coils. Here are the complete instructions. Because of the trans-
former and the low voltage it is absolutely safe.

   	How to construct a cheap under-gravel heater

Recently more and more reports have surfaced claiming that superior plant 
growing conditions and good algae control in an aquarium can be achieved if 
under-gravel heating coils are used.
  There are basically two competing systems on the market which both
promise good results. The first method uses a high-wattage heating coil
(0.5W/l = 100 Watt coil for 200 l tank) and is marketed mainly by Dupla.
This method requires a temperature regulator in order to limit the heat
build-up in the gravel. The coil functions as the basic source of heat.

  The other method, used by Dennerle, proposes low-wattage coils (0.1 - 0.2W/l
= 20 - 40 Watt coil for 200 l tank). In this case, additional heating by
a submersible heater is required to maintain tropical water temperatures 
(unless the room temperature is relatively high and stable). 
While Dennerle suggests and sells a temperature controller that controls 
both the main heater and the low wattage heating cable (the cable is treated 
with priority and will stay switched on as long as possible), it is not 
strictly necessary.  The low wattage makes it possible to leave the coil 
on permanently for most applications. (One exception might be non-air-
conditioned rooms, which get very hot in the summer; in that case one would 
probably need a chiller anyway).

  Anybody with some basic knowledge of electrical wiring can set up an 
under-gravel heating system, which is relatively inexpensive. Just follow 
the steps outlined below. 

WARNING: If you are not sure about your capabilities in electrical
wiring, have some knowledgeable person help you or don't do this!

1) First determine the wattage and the length of cable that you need.
A rule of thumb is 1-2W per 10 liters of water (0.4W/gal - 0.8W/gal).
For a tank of 100cm x 50cm x 50cm we have 10x5x5 = 250 l gross volume. 
Let's say we deduct 20 l for thickness of the glass and another 50 l
for substrate volume. That leaves 180 l water. We choose 2W/10 l and
thus the required heating power is 18*2 = 36 Watts.
You also need to determine the approximate length of the cable. Normally
the cable is meandered on the tank bottom like this:

------------------------------------------------------------------------   ^
|A                                                                     |   |
|++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++\   |
|                                                                   )  |
| /++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++/   |   |
|(                                                                     |
| \++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++\   |   D
|                                                                   )  |
| /++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++/   |   |
|(                                                                     |
| \+++++++++++++++++++++++++++++++++^++++++++++++++++++++++++++++++\   |   |
|                                   X                               )  |
|+++++++++++++++++++++++++++++++++++v++++++++++++++++++++++++++++++/   |   |
|B                                                                     |
------------------------------------------------------------------------   v
< -  -   -   -   -   -   -  W -   -   -   -   -   -   -   -   -   -   ->

The feeding cable to the transformer is connected at points A and B.
Of course you can choose a slightly different arrangement: have point B end 
up at the same location as point A, for example. 
The important things to watch are:

     - have your cable go the entire length of the tank.
     - make sure the spacing X between the parallel lengths of cable is
       between 5 - 10 cm (2 - 4"), not any closer or wider.
     - try to avoid having one part of the heating cable touch another part.
       Although a short under the gravel is harmless (you won't get
       zapped) is is impossible to repair without re-doing your tank.
       Only if you can't get the right size wire would I suggest
       folding the wire in half so that you effectively double the
       wire length. That should rarely ever be necessary.

The length is obviously D + n*W. n is defined as  D/X.  {(D-X)/X + 1} 
Example: tank has bottom surface area of 100cm x 50 cm. Spacing X is 7 cm;
it follows that n=50/7 = 7, length = 50 cm + 7*100 cm = 750 cm = 7.5m

2) The commonly available voltages for transformers are 6V, 9V, 12V, 15V and 
24V. Since P*R = V*V, we have the following table (P is 36W from the above
	  R = V^2/P                           |for 7.5m:
   V		V^2	R/Ohms 		I/A   |	Ohms/m
					      |		    wire gets thinner
   6		36	1		6     |	0.133		|
   9		81	2.25		4     |	0.300		|
   12		144	4		3     |	0.533		v

In the last column we calculated the resistance/length value a wire would 
need to have if it was 7.5m long (as in our example). We did that by dividing
column 3 (R/Ohms) by 7.5 (length of cable).
Now we need to find out which AWG wire size has the appropriate resistance
per meter. The following table helps:

   AWG		Dia/mm		Ohms/m		Ohms/1000ft
   26		0.405		0.136		41.62        <--- 6V
   27		0.361		0.172		52.48
   28		0.321		0.217		66.17
   29		0.286		0.274		83.44      \
   30		0.255		0.345		105.2      / ++++++ 9V
   31		0.227		0.435		132.7
   32		0.202		0.548		167.3        <---- 12V
   33		0.180		0.692		211
   34		0.160		0.873		266

As you can see the 6V transformer would require AWG 26 wire (7.5m) and
the 12V transformer AWG 32 wire. For 9V there is not really a suitable
wire size. Keep in mind that we can tolerate some variation here. A 
value within 20% of that calculated would be fine. (Our initial rule of 
thumb called for 1 - 2W for 10 l, which is 1.5W +/- 30%.)
  It is always desirable to achieve the heating power with a low current
(= high voltage). In our 6V example 6 Amps of current were needed for 
the 36 Watts. This would require the feeding cables from the transformer 
to the heating cable to be relatively thick. For up to 10 Amps you can 
safely use AWG 13 or AWG 12.  Use a sufficient length of wire to reach 
from your aquarium substrate to the transformer.
For 12V only 3 Amps are needed. That would be the better solution IF 
AWG 32 wire is available. 
The cable insulation should be PVC or silicon and be able to withstand at 
least 80 C, better 100 C (most cables do that anyway). The cables 
(heating and feeding) must be spliced and soldered together and then 
insulated with heat shrink tubing. Silicon aquarium sealer could also be
used to insulate the bare metal from water contact. 

3) Select the transformer. There are three requirements for this.
First and most importantly the transformer must provide insulation
between the primary and secondary windings. The best type is the one that 
uses separate bobbins for primary and secondary windings. In that case there 
is no way that they could accidentally touch. 

  >> The transformer is the part your safety depends on! <<

Second, make sure the transformer will be able to handle the needed power
(36 Watts in our case). Buy a transformer that will handle a
20-30% higher wattage than you need. Otherwise it will get relatively warm.
Third, the current rating of the secondary winding must be equal to or higher 
than the current you calculated.
  Don't forget to put a fuse (slow blow) into the primary side when you wire 
everything up. The value of the fuse can be calculated: 
   I = P/120V (In our case 36W/120V = 0.3A = 300mA). 
Use a fuse with approximately twice that calculated value.

Here are examples of some prices based on German electronics mail order 
catalogs; they should be equivalent elsewhere:

AWG32 wire: 		$2 for 10m
AWG12 wire: 		$2.50 for 6m
heat shrink tubing:	$2 for a large assortment of all sizes
transformer 12V/4.17A:  $17
plastic enclosure       $6
small items (fuse etc.) $2
Total:		       $31.50

Under-gravel heating shows the best results if used with a substrate
like clay or tropical laterite (I have heard clay works quite well).
DON'T use anything organic (potting soil or peat).
  If you are using the low heat method you should also provide adequate
insulation for the bottom of your tank. Especially if you use a glass
tank the heat loss can be substantial. Take a 1cm thick sheet of styrofoam,
cut it to size and put it underneath the tank. This will also help eliminate
any mechanical tensions caused by an uneven surface of your tank support.