heater problems

[lgb]

Just for info.

I've seen the Neptune brand heater suggested as one of the better ones.
I may have gotten into a bad batch, but that hasn't been my experience.

I put a 25 watt one in a 10 gallon tank in November. In January it
stuck in the on position. I took it back to PetsMart and they replaced
it. Last weekend the second one failed as well.

In both cases I caught the temperature rise before boiling the fish, but
now I'm getting paranoid about heaters. I rechecked the manual to make
sure I wasn't doing anything wrong, and apparently I wasn't.

The only possibility I saw was the the heater should not touch bottom or
decorations. Mine didn't but they are in close proximity to some sag
plants and I assume the leaves touched the heaters from time to time in
the current. But this must be a common occurence in planted tanks.

BTW, I've got a cheap heater from Big Al's on another planted tank and
it's still doing fine after 5 months.

If anyone knows why a supposedly reliable brand has been giving me
trouble, please let me know.

--
Homo sapiens is a goal, not a description

[bettasngoldfish]

The reason you may be having problems with your heater is that you have
the wrong size heater. I use the 25 watt Neptunes in my 2 1/2 gallons
tanks. In a 10 gallon tank you should really have a 50 watt heater. I
have three Neptunes and have never had a problem with them.

Maria

[lgb]

In article . com>,
says...
> The reason you may be having problems with your heater is that you have
> the wrong size heater. I use the 25 watt Neptunes in my 2 1/2 gallons
> tanks. In a 10 gallon tank you should really have a 50 watt heater. I
> have three Neptunes and have never had a problem with them.
>
Well, the Baensch (sp?) Aquarium Atlas suggests 2-3 watts per gallon. I
realize most say 10, but the reasoning is that 2-3 won't boil the water
if they stick open. And in a climate-controlled house, I haven't seen
the Neptunes run very often.

Not saying you're wrong, just that there are different opinions on
wattage per gallon.

--
Homo sapiens is a goal, not a description

[bettasngoldfish]

That's cool, everyone has their own opinions ; )

[Jim Anderson]

In article >,
says...
> In article . com>,
> says...
> > The reason you may be having problems with your heater is that you have
> > the wrong size heater. I use the 25 watt Neptunes in my 2 1/2 gallons
> > tanks. In a 10 gallon tank you should really have a 50 watt heater. I
> > have three Neptunes and have never had a problem with them.
> >
> Well, the Baensch (sp?) Aquarium Atlas suggests 2-3 watts per gallon. I
> realize most say 10, but the reasoning is that 2-3 won't boil the water
> if they stick open. And in a climate-controlled house, I haven't seen
> the Neptunes run very often.
>
> Not saying you're wrong, just that there are different opinions on
> wattage per gallon.
>
>

"2.5 watts will keep 1 gallon of water 8-10 degrees F above room temp."
Is what I have in my notes.


--
Jim Anderson
( 8(|) To email me just pull my_finger

[Dick]

On Tue, 5 Apr 2005 16:56:37 -0700, lgb > wrote:

>In article . com>,
says...
>> The reason you may be having problems with your heater is that you have
>> the wrong size heater. I use the 25 watt Neptunes in my 2 1/2 gallons
>> tanks. In a 10 gallon tank you should really have a 50 watt heater. I
>> have three Neptunes and have never had a problem with them.
>>
>Well, the Baensch (sp?) Aquarium Atlas suggests 2-3 watts per gallon. I
>realize most say 10, but the reasoning is that 2-3 won't boil the water
>if they stick open. And in a climate-controlled house, I haven't seen
>the Neptunes run very often.
>
>Not saying you're wrong, just that there are different opinions on
>wattage per gallon.

I am surprised that none of the so called formulas consider the
difference between the desired tank temperature and the room
temperature. In the summer my room averages 75 degrees, and tank 77.
I don't need much heating at all. My utility room, in the winter, is
only in the 60's, plus when the outside door is open, it has a
temporary drop. I have a 10 gallon quarantine tank that I keep
occupied in that room. I use 150 watts, but I keep an electronic
thermometer with low/hi alarm to monitor the tank. A room with a
steady temperature needs less tank heat than one that varies more
widely.

One problem I have with the low watt formula if there is much
temperature gradient is the need for the heater to run more. Most of
the thermostats in the heaters are mechanical. The more they cycle,
the more chance for the contacts to pit.

I like the electronic heaters and rather have too much heat than too
little.

Whatever you choose to do, be sure you are comfortable with your
answer. Nothing takes away pleasure than to be worried that your
answer is a safe one.

dick

[NetMax]

"Dick" > wrote in message
...
> On Tue, 5 Apr 2005 16:56:37 -0700, lgb > wrote:
>
>>In article . com>,
says...
>>> The reason you may be having problems with your heater is that you
>>> have
>>> the wrong size heater. I use the 25 watt Neptunes in my 2 1/2 gallons
>>> tanks. In a 10 gallon tank you should really have a 50 watt heater.
>>> I
>>> have three Neptunes and have never had a problem with them.
>>>
>>Well, the Baensch (sp?) Aquarium Atlas suggests 2-3 watts per gallon.
>>I
>>realize most say 10, but the reasoning is that 2-3 won't boil the water
>>if they stick open. And in a climate-controlled house, I haven't seen
>>the Neptunes run very often.
>>
>>Not saying you're wrong, just that there are different opinions on
>>wattage per gallon.
>
> I am surprised that none of the so called formulas consider the
> difference between the desired tank temperature and the room
> temperature. In the summer my room averages 75 degrees, and tank 77.
> I don't need much heating at all. My utility room, in the winter, is
> only in the 60's, plus when the outside door is open, it has a
> temporary drop. I have a 10 gallon quarantine tank that I keep
> occupied in that room. I use 150 watts, but I keep an electronic
> thermometer with low/hi alarm to monitor the tank. A room with a
> steady temperature needs less tank heat than one that varies more
> widely.
>
> One problem I have with the low watt formula if there is much
> temperature gradient is the need for the heater to run more. Most of
> the thermostats in the heaters are mechanical. The more they cycle,
> the more chance for the contacts to pit.
>
> I like the electronic heaters and rather have too much heat than too
> little.
>
> Whatever you choose to do, be sure you are comfortable with your
> answer. Nothing takes away pleasure than to be worried that your
> answer is a safe one.
>
> dick

Right you are. What I use is 2-3wpg for upstairs tanks and 4-5wpg for
downstairs tanks. Basement tanks almost always need more heat than a tank
on the top floor of a house. Adjust the wpg up and down by the part of
the world you're in (using the largest temperature difference you expect
to get and leave yourself some margin).
--
www.NetMax.tk

[lgb]

> > One problem I have with the low watt formula if there is much
> > temperature gradient is the need for the heater to run more. Most of
> > the thermostats in the heaters are mechanical. The more they cycle,
> > the more chance for the contacts to pit.
>

I missed this the first time. AFAIK, the amount of cycling is
controlled by the rate at which the water loses heat. The only thing
the watts control is the amount of time it takes to restore the lost
heat.

So the cycling rate should be the same for any heater. Am I wrong?

--
Homo sapiens is a goal, not a description

[NetMax]

"lgb" > wrote in message
...
>
>> > One problem I have with the low watt formula if there is much
>> > temperature gradient is the need for the heater to run more. Most
>> > of
>> > the thermostats in the heaters are mechanical. The more they cycle,
>> > the more chance for the contacts to pit.
>>
>
> I missed this the first time. AFAIK, the amount of cycling is
> controlled by the rate at which the water loses heat. The only thing
> the watts control is the amount of time it takes to restore the lost
> heat.
>
> So the cycling rate should be the same for any heater. Am I wrong?
>

Partially wrong. Watts *are* the heat capacity available, but the on/off
cycling of a heater depends on more things. The heater will not go on
when the water is just below the set point, or it would be constantly
cycling on and off (especially as its own heat would be fooling its
internal thermostat. Heaters are designed to have 'hysterisis' which is
a window of operation. For example, if set to 77F, the heater might go
on at 76 and off at 77.5, so it has a window of 1.5F. Manufacturers set
different windows, and the adjusted window will vary by unit. Too big a
window and people will notice that the temperature varies a bit, and too
narrow a window and analog contacts will fail much earlier.

The 2nd point is that not all heaters operate a set of contacts which
might pit. Digital heaters commonly replace these analog contacts with
digital switches, though I'm not certain what type is commonly used, SCRs
to trim the ac, FETs to feather the voltage, or just some type of an
on/off device to replace the functionality of the analog contacts. In
any case, there is no mechanical part to wear out (except for the heating
coil which eventually break from the thermal stress, just like the
filament in an incandescent light bulb). hth
--
www.NetMax.tk

[Dick]

On Thu, 7 Apr 2005 09:20:41 -0700, lgb > wrote:

>
>> > One problem I have with the low watt formula if there is much
>> > temperature gradient is the need for the heater to run more. Most of
>> > the thermostats in the heaters are mechanical. The more they cycle,
>> > the more chance for the contacts to pit.
>>
>
>I missed this the first time. AFAIK, the amount of cycling is
>controlled by the rate at which the water loses heat. The only thing
>the watts control is the amount of time it takes to restore the lost
>heat.
>
>So the cycling rate should be the same for any heater. Am I wrong?

I think you are right. I would guess low watts would cause the
contacts to be in touch for long periods with the possibility that the
tank temperature could drop before the low watts could get it backup,
whereas the high watts would be together for shorter periods and have
the reserve to keep the tank warm even if there is a large drop in
room temperature..

Frequent cycling may be more related to tank size. A larger tank,
once heated, retains more heat and would be slower to drop its
temperature, where as the smaller mass of water in small tank reacts
quicker to room temperature changes. Just guessing, but sounds
reasonable.

I wondered about what I had said regarding low watts and pitting. I
am not sure about cause and effect, however, having said that, the two
common mechanical problems remain: pitting and sticking. I am not
sure if the two are related. Some people take their heaters apart and
sand the contacts to smooth their surfaces.

I prefer to spend the extra and get electronic thermostats with the
heater. I like the extra watts to know that if the room temperature
drops, the tanks will stay warm. As a precaution, I keep a digital
thermometer/hi-low alarm in each tank.

I am glad to know you read what I said carefully enough and was
willing to question my mistake.

dick

[Dick]

On Thu, 7 Apr 2005 23:04:19 -0400, "NetMax"
> wrote:

>"lgb" > wrote in message
...
>>
>>> > One problem I have with the low watt formula if there is much
>>> > temperature gradient is the need for the heater to run more. Most
>>> > of
>>> > the thermostats in the heaters are mechanical. The more they cycle,
>>> > the more chance for the contacts to pit.
>>>
>>
>> I missed this the first time. AFAIK, the amount of cycling is
>> controlled by the rate at which the water loses heat. The only thing
>> the watts control is the amount of time it takes to restore the lost
>> heat.
>>
>> So the cycling rate should be the same for any heater. Am I wrong?
>>
>
>Partially wrong. Watts *are* the heat capacity available, but the on/off
>cycling of a heater depends on more things. The heater will not go on
>when the water is just below the set point, or it would be constantly
>cycling on and off (especially as its own heat would be fooling its
>internal thermostat. Heaters are designed to have 'hysterisis' which is
>a window of operation. For example, if set to 77F, the heater might go
>on at 76 and off at 77.5, so it has a window of 1.5F. Manufacturers set
>different windows, and the adjusted window will vary by unit. Too big a
>window and people will notice that the temperature varies a bit, and too
>narrow a window and analog contacts will fail much earlier.
>
>The 2nd point is that not all heaters operate a set of contacts which
>might pit. Digital heaters commonly replace these analog contacts with
>digital switches, though I'm not certain what type is commonly used, SCRs
>to trim the ac, FETs to feather the voltage, or just some type of an
>on/off device to replace the functionality of the analog contacts. In
>any case, there is no mechanical part to wear out (except for the heating
>coil which eventually break from the thermal stress, just like the
>filament in an incandescent light bulb). hth


Hi Netmax,

I just hazard another guess regarding frequency of cycling. I said
that a heater is going to cycle more often in smaller tanks since the
mass loses its stored heat more quickly, but gains heat more quickly.

I had forgotten that not all thermostats have the same "hysteresis".
I would use a different word, but for the life of me I can't remember
what it is. I sure this will bug me today until the word I want
floats by. "Tolerance" is one word that comes to mind as the measure
of hysteresis. A tight tolerance would lead to more frequent cycling,
right?

This has been an interesting discussion. I did what I did to heat and
protect my tanks over 2 years ago and only had one occasions to fix
them (one heater stuck closed, not a digital, but now is).

It is amazing that so many people, new to the hobby, are faced with
such important questions when they start. The internet provides lots
of answers, but first one needs to know the questions.

Even though these newsgroups offer sometimes differing solutions at
least we present the questions and through the give and take indicate
thoughts to pursue. Once the questions are known, Google is useful.

dick

[lgb]

In article >,
says...
> I had forgotten that not all thermostats have the same "hysteresis".
> I would use a different word, but for the life of me I can't remember
> what it is. I sure this will bug me today until the word I want
> floats by.
>
Anticipator? That's what the old mercury furnace thermostats called it.

--
Homo sapiens is a goal, not a description

[lgb]

In article >,
says...
> The 2nd point is that not all heaters operate a set of contacts which
> might pit. Digital heaters commonly replace these analog contacts with
> digital switches, though I'm not certain what type is commonly used,
>

Sounds good. I'll look at that next time. I suspect they may be too
expensive for my budget, but I hope I'm wrong.

--
Homo sapiens is a goal, not a description

[NetMax]

"Dick" > wrote in message
...
> On Thu, 7 Apr 2005 23:04:19 -0400, "NetMax"
> > wrote:
>
>>"lgb" > wrote in message
...
>>>
>>>> > One problem I have with the low watt formula if there is much
>>>> > temperature gradient is the need for the heater to run more. Most
>>>> > of
>>>> > the thermostats in the heaters are mechanical. The more they
>>>> > cycle,
>>>> > the more chance for the contacts to pit.
>>>>
>>>
>>> I missed this the first time. AFAIK, the amount of cycling is
>>> controlled by the rate at which the water loses heat. The only thing
>>> the watts control is the amount of time it takes to restore the lost
>>> heat.
>>>
>>> So the cycling rate should be the same for any heater. Am I wrong?
>>>
>>
>>Partially wrong. Watts *are* the heat capacity available, but the
>>on/off
>>cycling of a heater depends on more things. The heater will not go on
>>when the water is just below the set point, or it would be constantly
>>cycling on and off (especially as its own heat would be fooling its
>>internal thermostat. Heaters are designed to have 'hysterisis' which
>>is
>>a window of operation. For example, if set to 77F, the heater might go
>>on at 76 and off at 77.5, so it has a window of 1.5F. Manufacturers
>>set
>>different windows, and the adjusted window will vary by unit. Too big
>>a
>>window and people will notice that the temperature varies a bit, and
>>too
>>narrow a window and analog contacts will fail much earlier.
>>
>>The 2nd point is that not all heaters operate a set of contacts which
>>might pit. Digital heaters commonly replace these analog contacts with
>>digital switches, though I'm not certain what type is commonly used,
>>SCRs
>>to trim the ac, FETs to feather the voltage, or just some type of an
>>on/off device to replace the functionality of the analog contacts. In
>>any case, there is no mechanical part to wear out (except for the
>>heating
>>coil which eventually break from the thermal stress, just like the
>>filament in an incandescent light bulb). hth
>
>
> Hi Netmax,
>
> I just hazard another guess regarding frequency of cycling. I said
> that a heater is going to cycle more often in smaller tanks since the
> mass loses its stored heat more quickly, but gains heat more quickly.

In theory or in practise? In theory, no, the cycle interval is more
closely tied with the ratio of heater's wattage to the rate of heat loss.
For any given size of tank, if the heater is sized proportionally smaller
or larger, its ON time to cause a temperature change will be the same, so
the cycle interval should be the same.

In practise, heater sizes are in relatively large 25W steps (relative to
a small tank) and they start high (ie: 50W), so small tanks tend to be
overheated (more than the recommended 3 to 5 wpg), so they will heat
faster, and cycle faster.

Even if the heater was sized exactly, smaller tanks have a much higher
glass to water ratio than large tanks, making them much more prone to
room temperature extremes, so the heaters have to work a lot harder.
This means that you need to start with a slightly more powerful heater
than needed, and the heater will tend to cycle more often because of that
(so you are right again ;~).

> I had forgotten that not all thermostats have the same "hysteresis".
> I would use a different word, but for the life of me I can't remember
> what it is. I sure this will bug me today until the word I want
> floats by. "Tolerance" is one word that comes to mind as the measure
> of hysteresis. A tight tolerance would lead to more frequent cycling,
> right?

Generally speaking, 'tolerance' applies to how tightly a value can be
obtained or maintained, so in your example a tight tolerance would indeed
cause the heater to cycle more.

I'm not sure what word has escaped you. In terms of environmental
controls, with analog designs, they do things to delay ON or delay OFF to
widen the hysterisis. In the digital world, there is almost no limit,
including using history, rate of change to adjust the hysteris, dual
sensors to overdrive the air (or heat transfer fluid) to more quickly
change the temperature of the device being controlled etc etc. I had the
opportunity to play in environmental engineering for a few years, and it
was quite an eye opener, with a lot of neat stuff which is applicable to
everyday things.

> This has been an interesting discussion. I did what I did to heat and
> protect my tanks over 2 years ago and only had one occasions to fix
> them (one heater stuck closed, not a digital, but now is).

I don't own a heater which I haven't at some point had to repair, and
I'll confess that this includes several submersible heaters which are not
really supposed to be servicable ;~). Armed with that experience, I
wouldn't recommend it to anyone.
--
www.NetMax.tk

> It is amazing that so many people, new to the hobby, are faced with
> such important questions when they start. The internet provides lots
> of answers, but first one needs to know the questions.
>
> Even though these newsgroups offer sometimes differing solutions at
> least we present the questions and through the give and take indicate
> thoughts to pursue. Once the questions are known, Google is useful.
>
> dick

[NetMax]

"Dick" > wrote in message
...
> On Thu, 7 Apr 2005 09:20:41 -0700, lgb > wrote:
>
>>
>>> > One problem I have with the low watt formula if there is much
>>> > temperature gradient is the need for the heater to run more. Most
>>> > of
>>> > the thermostats in the heaters are mechanical. The more they
>>> > cycle,
>>> > the more chance for the contacts to pit.
>>>
>>
>>I missed this the first time. AFAIK, the amount of cycling is
>>controlled by the rate at which the water loses heat. The only thing
>>the watts control is the amount of time it takes to restore the lost
>>heat.
>>
>>So the cycling rate should be the same for any heater. Am I wrong?
>
> I think you are right. I would guess low watts would cause the
> contacts to be in touch for long periods with the possibility that the
> tank temperature could drop before the low watts could get it backup,
> whereas the high watts would be together for shorter periods and have
> the reserve to keep the tank warm even if there is a large drop in
> room temperature..
>
> Frequent cycling may be more related to tank size. A larger tank,
> once heated, retains more heat and would be slower to drop its
> temperature, where as the smaller mass of water in small tank reacts
> quicker to room temperature changes. Just guessing, but sounds
> reasonable.
>
> I wondered about what I had said regarding low watts and pitting. I
> am not sure about cause and effect, however, having said that, the two
> common mechanical problems remain: pitting and sticking. I am not
> sure if the two are related. Some people take their heaters apart and
> sand the contacts to smooth their surfaces.
>
> I prefer to spend the extra and get electronic thermostats with the
> heater. I like the extra watts to know that if the room temperature
> drops, the tanks will stay warm. As a precaution, I keep a digital
> thermometer/hi-low alarm in each tank.
>
> I am glad to know you read what I said carefully enough and was
> willing to question my mistake.
>
> dick


In regards to the probability of pitting and sticking occurring more (or
less) with smaller heaters, I had to mull it over a bit but I think I've
figured it out (but please correct me if anyone thinks I'm on the wrong
track). Warning, this is going to get a little technical (you can always
jump to the end ;~).

At first I thought, smaller heater = less current = smaller arc (less
pitting), but I think that might be so over-simplified as to be
incorrect. The pitting occurs when there is an arc (melts the point of
the arc's entry), and an arc occurs when the air gap is small enough for
the voltage potential to cause the electrons to jump across, and this
occurs twice per heater cycle (when the heater switches ON and then
later, OFF).

When turning ON, the arc doesn't see the heater's high coil resistance so
the arc starts between a potentially high voltage and ground (only after
there is current flow does a voltage develop and the current drops to be
a function of resistance). So when turning ON, there should be no
difference in the arc between small or large heaters. The arc's
intensity might depend more on the phase of the ac signal when it arcs
(ac 'alternates' or goes up and down). About every 8 to 10ms, there is
no voltage potential on a home's ac lines, so if the critical air gap
occurs just before this happens, there would probably be no arc (assumes
many things, but none are affected by the heater's size).

When the heater turns OFF, the arc's intensity is again a function of
phase, air gap, voltage potential and one new item, inductance. Now the
smaller heater's higher resistance would reduce the voltage potential
between the two points (= smaller arc), and it is interrupting a smaller
current (= smaller arc). In both cases (small heater and large heater)
the inductance 'pulls' the current and I think that a smaller heater has
a lower inductance (= smaller arc), (there are many variables, most I
can't remember the formulas to, but for a similar voltage, inductance is
a function of current, so I think I'm ok).

So - turning ON, probably no difference, and turning OFF, if there is a
difference, the smaller heater has the advantage. This assumes they are
all cycling at the same rate. Also note that contact 'sticking' occurs
when contacts close (their pits line up and reflow together, 'gluing'
them together), so turning ON might be the more critical operation, (so
the smaller heater has just lost all of its advantages ;~).
--
www.NetMax.tk

[Dick]

On Fri, 8 Apr 2005 08:36:00 -0700, lgb > wrote:

>In article >,
says...
>> I had forgotten that not all thermostats have the same "hysteresis".
>> I would use a different word, but for the life of me I can't remember
>> what it is. I sure this will bug me today until the word I want
>> floats by.
>>
>Anticipator? That's what the old mercury furnace thermostats called it.

Nope, the anticipator is a small heater near the thermostat that makes
the thermostat believe the room is warmer than it is, so the
thermostat opens and starts the off cycle. Without the anticipator,
the room temperature would get too hot as it taes a while for the fan
to cool the plenum (the place where the fire actually heats). So with
pre heating the thermostat, the contacts separate when the room
temperature is slightly cooler than the thermostat setting turning off
the heat, knowing that the plenum heat will continue to heat the room.
Without the anticipator, the room would always get hotter than what is
set at the thermostat.

What I am thinking of describes how close to the temperature the
heater holds the tank. A loose tolerance might be +/- 2 degree versus
a tight tolerance of +/- 1 degree which means a thermostat with a
loose tolerance which is set for 78f actually would not turn on until
the water temperature went down to 76f and wouldn't turn off until 80f
whereas the tight tolerance would turn on at 77f and off at 79f.

dick

[Dick]

On Fri, 8 Apr 2005 23:13:36 -0400, "NetMax"
> wrote:

>"Dick" > wrote in message
...
>> On Thu, 7 Apr 2005 23:04:19 -0400, "NetMax"
>> > wrote:
>>
>>>"lgb" > wrote in message
...
>>>>
>>>>> > One problem I have with the low watt formula if there is much
>>>>> > temperature gradient is the need for the heater to run more. Most
>>>>> > of
>>>>> > the thermostats in the heaters are mechanical. The more they
>>>>> > cycle,
>>>>> > the more chance for the contacts to pit.
>>>>>
>>>>
>>>> I missed this the first time. AFAIK, the amount of cycling is
>>>> controlled by the rate at which the water loses heat. The only thing
>>>> the watts control is the amount of time it takes to restore the lost
>>>> heat.
>>>>
>>>> So the cycling rate should be the same for any heater. Am I wrong?
>>>>
>>>
>>>Partially wrong. Watts *are* the heat capacity available, but the
>>>on/off
>>>cycling of a heater depends on more things. The heater will not go on
>>>when the water is just below the set point, or it would be constantly
>>>cycling on and off (especially as its own heat would be fooling its
>>>internal thermostat. Heaters are designed to have 'hysterisis' which
>>>is
>>>a window of operation. For example, if set to 77F, the heater might go
>>>on at 76 and off at 77.5, so it has a window of 1.5F. Manufacturers
>>>set
>>>different windows, and the adjusted window will vary by unit. Too big
>>>a
>>>window and people will notice that the temperature varies a bit, and
>>>too
>>>narrow a window and analog contacts will fail much earlier.
>>>
>>>The 2nd point is that not all heaters operate a set of contacts which
>>>might pit. Digital heaters commonly replace these analog contacts with
>>>digital switches, though I'm not certain what type is commonly used,
>>>SCRs
>>>to trim the ac, FETs to feather the voltage, or just some type of an
>>>on/off device to replace the functionality of the analog contacts. In
>>>any case, there is no mechanical part to wear out (except for the
>>>heating
>>>coil which eventually break from the thermal stress, just like the
>>>filament in an incandescent light bulb). hth
>>
>>
>> Hi Netmax,
>>
>> I just hazard another guess regarding frequency of cycling. I said
>> that a heater is going to cycle more often in smaller tanks since the
>> mass loses its stored heat more quickly, but gains heat more quickly.
>
>In theory or in practise? In theory, no, the cycle interval is more
>closely tied with the ratio of heater's wattage to the rate of heat loss.
>For any given size of tank, if the heater is sized proportionally smaller
>or larger, its ON time to cause a temperature change will be the same, so
>the cycle interval should be the same.
>
>In practise, heater sizes are in relatively large 25W steps (relative to
>a small tank) and they start high (ie: 50W), so small tanks tend to be
>overheated (more than the recommended 3 to 5 wpg), so they will heat
>faster, and cycle faster.
>
>Even if the heater was sized exactly, smaller tanks have a much higher
>glass to water ratio than large tanks, making them much more prone to
>room temperature extremes, so the heaters have to work a lot harder.
>This means that you need to start with a slightly more powerful heater
>than needed, and the heater will tend to cycle more often because of that
>(so you are right again ;~).
>
>> I had forgotten that not all thermostats have the same "hysteresis".
>> I would use a different word, but for the life of me I can't remember
>> what it is. I sure this will bug me today until the word I want
>> floats by. "Tolerance" is one word that comes to mind as the measure
>> of hysteresis. A tight tolerance would lead to more frequent cycling,
>> right?
>
>Generally speaking, 'tolerance' applies to how tightly a value can be
>obtained or maintained, so in your example a tight tolerance would indeed
>cause the heater to cycle more.
>
>I'm not sure what word has escaped you. In terms of environmental
>controls, with analog designs, they do things to delay ON or delay OFF to
>widen the hysterisis. In the digital world, there is almost no limit,
>including using history, rate of change to adjust the hysteris, dual
>sensors to overdrive the air (or heat transfer fluid) to more quickly
>change the temperature of the device being controlled etc etc. I had the
>opportunity to play in environmental engineering for a few years, and it
>was quite an eye opener, with a lot of neat stuff which is applicable to
>everyday things.
>
>> This has been an interesting discussion. I did what I did to heat and
>> protect my tanks over 2 years ago and only had one occasions to fix
>> them (one heater stuck closed, not a digital, but now is).
>
>I don't own a heater which I haven't at some point had to repair, and
>I'll confess that this includes several submersible heaters which are not
>really supposed to be servicable ;~). Armed with that experience, I
>wouldn't recommend it to anyone.

I thought of another factor which we haven't yet mentioned effecting
the heating cycle: location. Aside from the obvious overheating from
direct sunlight, I am thinking of what wall the tank sets near,
outside and inside. My house has adobe walls and take longer to
change temperature than do most house materials, so wall mass is
another consideration.

An outside wall is cooler and absorbs heat from the tank faster than
an inside wall which stays closer to the room air temperature. On the
outside wall heat from the tank is losing heat to the cooler outside
wall temperature. In my case I have three tanks set by outside walls
and 2 near inside walls. The inside wall temperatures stays close to
the room temperature, whereas the wall temperature are 2 degrees
cooler. It would ease the burden on the tank temperature for the
tanks to be located near an inside wall.

dick

[NetMax]

"Dick" > wrote in message
...
> On Fri, 8 Apr 2005 23:13:36 -0400, "NetMax"
> > wrote:
>
>>"Dick" > wrote in message
...
>>> On Thu, 7 Apr 2005 23:04:19 -0400, "NetMax"
>>> > wrote:
>>>
>>>>"lgb" > wrote in message
...
>>>>>
>>>>>> > One problem I have with the low watt formula if there is much
>>>>>> > temperature gradient is the need for the heater to run more.
>>>>>> > Most
>>>>>> > of
>>>>>> > the thermostats in the heaters are mechanical. The more they
>>>>>> > cycle,
>>>>>> > the more chance for the contacts to pit.
>>>>>>
>>>>>
>>>>> I missed this the first time. AFAIK, the amount of cycling is
>>>>> controlled by the rate at which the water loses heat. The only
>>>>> thing
>>>>> the watts control is the amount of time it takes to restore the
>>>>> lost
>>>>> heat.
>>>>>
>>>>> So the cycling rate should be the same for any heater. Am I wrong?
>>>>>
>>>>
>>>>Partially wrong. Watts *are* the heat capacity available, but the
>>>>on/off
>>>>cycling of a heater depends on more things. The heater will not go
>>>>on
>>>>when the water is just below the set point, or it would be constantly
>>>>cycling on and off (especially as its own heat would be fooling its
>>>>internal thermostat. Heaters are designed to have 'hysterisis' which
>>>>is
>>>>a window of operation. For example, if set to 77F, the heater might
>>>>go
>>>>on at 76 and off at 77.5, so it has a window of 1.5F. Manufacturers
>>>>set
>>>>different windows, and the adjusted window will vary by unit. Too
>>>>big
>>>>a
>>>>window and people will notice that the temperature varies a bit, and
>>>>too
>>>>narrow a window and analog contacts will fail much earlier.
>>>>
>>>>The 2nd point is that not all heaters operate a set of contacts which
>>>>might pit. Digital heaters commonly replace these analog contacts
>>>>with
>>>>digital switches, though I'm not certain what type is commonly used,
>>>>SCRs
>>>>to trim the ac, FETs to feather the voltage, or just some type of an
>>>>on/off device to replace the functionality of the analog contacts.
>>>>In
>>>>any case, there is no mechanical part to wear out (except for the
>>>>heating
>>>>coil which eventually break from the thermal stress, just like the
>>>>filament in an incandescent light bulb). hth
>>>
>>>
>>> Hi Netmax,
>>>
>>> I just hazard another guess regarding frequency of cycling. I said
>>> that a heater is going to cycle more often in smaller tanks since the
>>> mass loses its stored heat more quickly, but gains heat more quickly.
>>
>>In theory or in practise? In theory, no, the cycle interval is more
>>closely tied with the ratio of heater's wattage to the rate of heat
>>loss.
>>For any given size of tank, if the heater is sized proportionally
>>smaller
>>or larger, its ON time to cause a temperature change will be the same,
>>so
>>the cycle interval should be the same.
>>
>>In practise, heater sizes are in relatively large 25W steps (relative
>>to
>>a small tank) and they start high (ie: 50W), so small tanks tend to be
>>overheated (more than the recommended 3 to 5 wpg), so they will heat
>>faster, and cycle faster.
>>
>>Even if the heater was sized exactly, smaller tanks have a much higher
>>glass to water ratio than large tanks, making them much more prone to
>>room temperature extremes, so the heaters have to work a lot harder.
>>This means that you need to start with a slightly more powerful heater
>>than needed, and the heater will tend to cycle more often because of
>>that
>>(so you are right again ;~).
>>
>>> I had forgotten that not all thermostats have the same "hysteresis".
>>> I would use a different word, but for the life of me I can't remember
>>> what it is. I sure this will bug me today until the word I want
>>> floats by. "Tolerance" is one word that comes to mind as the measure
>>> of hysteresis. A tight tolerance would lead to more frequent
>>> cycling,
>>> right?
>>
>>Generally speaking, 'tolerance' applies to how tightly a value can be
>>obtained or maintained, so in your example a tight tolerance would
>>indeed
>>cause the heater to cycle more.
>>
>>I'm not sure what word has escaped you. In terms of environmental
>>controls, with analog designs, they do things to delay ON or delay OFF
>>to
>>widen the hysterisis. In the digital world, there is almost no limit,
>>including using history, rate of change to adjust the hysteris, dual
>>sensors to overdrive the air (or heat transfer fluid) to more quickly
>>change the temperature of the device being controlled etc etc. I had
>>the
>>opportunity to play in environmental engineering for a few years, and
>>it
>>was quite an eye opener, with a lot of neat stuff which is applicable
>>to
>>everyday things.
>>
>>> This has been an interesting discussion. I did what I did to heat
>>> and
>>> protect my tanks over 2 years ago and only had one occasions to fix
>>> them (one heater stuck closed, not a digital, but now is).
>>
>>I don't own a heater which I haven't at some point had to repair, and
>>I'll confess that this includes several submersible heaters which are
>>not
>>really supposed to be servicable ;~). Armed with that experience, I
>>wouldn't recommend it to anyone.
>
> I thought of another factor which we haven't yet mentioned effecting
> the heating cycle: location. Aside from the obvious overheating from
> direct sunlight, I am thinking of what wall the tank sets near,
> outside and inside. My house has adobe walls and take longer to
> change temperature than do most house materials, so wall mass is
> another consideration.
>
> An outside wall is cooler and absorbs heat from the tank faster than
> an inside wall which stays closer to the room air temperature. On the
> outside wall heat from the tank is losing heat to the cooler outside
> wall temperature. In my case I have three tanks set by outside walls
> and 2 near inside walls. The inside wall temperatures stays close to
> the room temperature, whereas the wall temperature are 2 degrees
> cooler. It would ease the burden on the tank temperature for the
> tanks to be located near an inside wall.
>
> dick

Sounds right. Outside wall installations would decrease the wattage to
losses ratio, so the heater would be ON more and longer, theoretically
cycling less often (typically). The other extreme is a very well
insulated tank in an air temperature close to the tank's water
temperature. In this scenario, the ratio of wattage to losses goes very
high, and the cycling interval *reverses* its trend and *decreases*.
Beyond a certain point, as the ratio reaches infinity, the interval
approaches zero (isn't science fun?).

On a more practical note, when I started installing tanks into
walls/bookcases, I noticed this external wall cooling effect much more
(the gap behind the tank didn't have any significant circulation and got
very cold. My solution was to put a sheet of black painted chipboard as
a backing behind the tank, and then I filled the air gap with some cheap
sponge material (which also pushed and held the chipboard against the
tank). I also keep a 2 or 3" PVC pipe in the sponge material (to run
hoses and wires from under the tank). I keep a long string attached at
both points (above and below the PVC pipe) to pull new wires/hoses up and
down through the PVC pipe.
--
www.NetMax.tk

[George Pontis]

In article >,
says...
>
> When the heater turns OFF, the arc's intensity is again a function of
> phase, air gap, voltage potential and one new item, inductance. Now the
> smaller heater's higher resistance would reduce the voltage potential
> between the two points (= smaller arc), and it is interrupting a smaller
> current (= smaller arc). In both cases (small heater and large heater)
> the inductance 'pulls' the current and I think that a smaller heater has
> a lower inductance (= smaller arc), (there are many variables, most I
> can't remember the formulas to, but for a similar voltage, inductance is
> a function of current, so I think I'm ok).
>
> So - turning ON, probably no difference, and turning OFF, if there is a
> difference, the smaller heater has the advantage. This assumes they are
> all cycling at the same rate. Also note that contact 'sticking' occurs
> when contacts close (their pits line up and reflow together, 'gluing'
> them together), so turning ON might be the more critical operation, (so
> the smaller heater has just lost all of its advantages ;~).
>

It turns out that the inductance of an aquarium heater is very small. At 60Hz
(even more so at 50 Hz), the resistance of the heater completely dominates the
inductance. I was going to measure an Acura 1000 (300W) to provide some real
numbers. Turns out I couldn't find the leads for the LCR meter ... but plugged it
into a watt/PF meter instead. The power factor was 1.00, confirming no inductive
component. Your memory of inductance being a function of current would apply to an
inductor with an iron core, or some other core made from a magnetic material that
could saturate. These heaters use a ceramic core which has no affect on the
inductance. So all we have in an aquarium heater is the long spiraled coil, and it
does not add up to enough inductance to play any role at AC power frequencies.

I never tried repairing a heater but I once took one apart. It had a magnet on the
bimetallic contact. No only would that add some hysteresis, but it helps the
contacts open and close briskly so the arc time is minimized.

Have you ever had to disassemble or repair an electronic heater like the Hagen
Tronic ? I always wondered what they used for a switch. I assume it is a triac,
but then how do they keep the heating of the triac from influencing the
temperature sensor ???

George

[NetMax]

"George Pontis" > wrote in message
...
> In article >,
> says...
>>
>> When the heater turns OFF, the arc's intensity is again a function of
>> phase, air gap, voltage potential and one new item, inductance. Now
>> the
>> smaller heater's higher resistance would reduce the voltage potential
>> between the two points (= smaller arc), and it is interrupting a
>> smaller
>> current (= smaller arc). In both cases (small heater and large
>> heater)
>> the inductance 'pulls' the current and I think that a smaller heater
>> has
>> a lower inductance (= smaller arc), (there are many variables, most I
>> can't remember the formulas to, but for a similar voltage, inductance
>> is
>> a function of current, so I think I'm ok).
>>
>> So - turning ON, probably no difference, and turning OFF, if there is
>> a
>> difference, the smaller heater has the advantage. This assumes they
>> are
>> all cycling at the same rate. Also note that contact 'sticking'
>> occurs
>> when contacts close (their pits line up and reflow together, 'gluing'
>> them together), so turning ON might be the more critical operation,
>> (so
>> the smaller heater has just lost all of its advantages ;~).
>>
>
> It turns out that the inductance of an aquarium heater is very small.
> At 60Hz
> (even more so at 50 Hz), the resistance of the heater completely
> dominates the
> inductance. I was going to measure an Acura 1000 (300W) to provide some
> real
> numbers. Turns out I couldn't find the leads for the LCR meter ... but
> plugged it
> into a watt/PF meter instead. The power factor was 1.00, confirming no
> inductive
> component. Your memory of inductance being a function of current would
> apply to an
> inductor with an iron core, or some other core made from a magnetic
> material that
> could saturate. These heaters use a ceramic core which has no affect on
> the
> inductance. So all we have in an aquarium heater is the long spiraled
> coil, and it
> does not add up to enough inductance to play any role at AC power
> frequencies.

Thanks George, that makes perfect sense. Not much in the way of dipoles
in a ceramic core ;~)

> I never tried repairing a heater but I once took one apart. It had a
> magnet on the
> bimetallic contact. No only would that add some hysteresis, but it
> helps the
> contacts open and close briskly so the arc time is minimized.

I've also seen high temperature plating used on the contacts (Thermal
heaters), but the magnet sounds like an interesting idea. It would be
tough to convince me that a magnet would not have aging characteristics
though.

> Have you ever had to disassemble or repair an electronic heater like
> the Hagen
> Tronic ? I always wondered what they used for a switch. I assume it is
> a triac,
> but then how do they keep the heating of the triac from influencing the
> temperature sensor ???
>
> George

I just went and rifled through my stash of heaters and found parts for a
Rena submersible. Unfortunately they used bimetal contacts (magnetic!)
instead of a solid state switch, so I'm no closer to answering your
question. I would guess that if they can make a thermostat insensitive
to the heater core's effects, then the small amount of heat from a triac
would be manageable (especially if they were triggering the triac at zero
phase, which I think would let it run quite cool, but my electronics is
getting rusty from lack of use).
--
www.NetMax.tk

[Dick]

On Sat, 9 Apr 2005 12:45:39 -0700, George Pontis >
wrote:

>In article >,
says...
>>
>> When the heater turns OFF, the arc's intensity is again a function of
>> phase, air gap, voltage potential and one new item, inductance. Now the
>> smaller heater's higher resistance would reduce the voltage potential
>> between the two points (= smaller arc), and it is interrupting a smaller
>> current (= smaller arc). In both cases (small heater and large heater)
>> the inductance 'pulls' the current and I think that a smaller heater has
>> a lower inductance (= smaller arc), (there are many variables, most I
>> can't remember the formulas to, but for a similar voltage, inductance is
>> a function of current, so I think I'm ok).
>>
>> So - turning ON, probably no difference, and turning OFF, if there is a
>> difference, the smaller heater has the advantage. This assumes they are
>> all cycling at the same rate. Also note that contact 'sticking' occurs
>> when contacts close (their pits line up and reflow together, 'gluing'
>> them together), so turning ON might be the more critical operation, (so
>> the smaller heater has just lost all of its advantages ;~).
>>
>
>It turns out that the inductance of an aquarium heater is very small. At 60Hz
>(even more so at 50 Hz), the resistance of the heater completely dominates the
>inductance. I was going to measure an Acura 1000 (300W) to provide some real
>numbers. Turns out I couldn't find the leads for the LCR meter ... but plugged it
>into a watt/PF meter instead. The power factor was 1.00, confirming no inductive
>component. Your memory of inductance being a function of current would apply to an
>inductor with an iron core, or some other core made from a magnetic material that
>could saturate. These heaters use a ceramic core which has no affect on the
>inductance. So all we have in an aquarium heater is the long spiraled coil, and it
>does not add up to enough inductance to play any role at AC power frequencies.
>
>I never tried repairing a heater but I once took one apart. It had a magnet on the
>bimetallic contact. No only would that add some hysteresis, but it helps the
>contacts open and close briskly so the arc time is minimized.
>
>Have you ever had to disassemble or repair an electronic heater like the Hagen
>Tronic ? I always wondered what they used for a switch. I assume it is a triac,
>but then how do they keep the heating of the triac from influencing the
>temperature sensor ???
>
>George

My electronic heaters use an external thermocouple to sense the tank
temperature with the heater in a separate tube and the control in a
third package. I presume the thermocouple voltage is compared to a
reference voltage adjusted by the dial and when two differ the device
closes the heater /power circuit.

dick

[NetMax]

"Dick" > wrote in message
...
> On Sat, 9 Apr 2005 12:45:39 -0700, George Pontis >
> wrote:
>
>>In article >,
says...
>>>
<snip>

>>Have you ever had to disassemble or repair an electronic heater like
>>the Hagen
>>Tronic ? I always wondered what they used for a switch. I assume it is
>>a triac,
>>but then how do they keep the heating of the triac from influencing the
>>temperature sensor ???
>>
>>George
>
> My electronic heaters use an external thermocouple to sense the tank
> temperature with the heater in a separate tube and the control in a
> third package. I presume the thermocouple voltage is compared to a
> reference voltage adjusted by the dial and when two differ the device
> closes the heater /power circuit.
>
> dick

Almost exactly (I think ;~). Thermocouple voltages are so low, that they
probably amplify it first, or run it through the reference table and
generate a new output to compare with your dialled setting.
Incidentally, if you can see the wire colours going to your thermocouple,
this will indicate the type used:
http://www.omega.com/techref/colorcodes.html
If the wires are the same colour, then I suspect it is not a thermocouple
but actually a thermistor (which is what I would suspect as thermistors
are much cheaper to use in temperature control circuits).
--
www.NetMax.tk

[[email protected]]

I have 5 Visitherm Stealth heaters. All worked great.