Running a 208V radiant ceiling heater from 240V

[John Sevinsky]

I've always wanted radiant ceiling panel heat in our basement, but
those panels are expensive. Recently, I found a supply of used panels
for a reasonable price, so I bought one to play with. (It's a Qmark
Aztec 375W panel.) Unfortunately, it is rated for 208V, and our house
has 240V. I hooked it up, and it worked. But, the surface temperature
in the middle of the panel reached a maximum temperature of 250F, which
might be beyond the safe operating range.

So, I'm wondering how I can reduce the power supplied to these panels.
Although this won't be perfect, I thought about some diode arrangement
that would supply power to a pair of these panels in parallel for one
half of the cycle and in series for the other half. So, they would be
getting 240V half of the time and 120V half of the time. That should
average out to less than their rated capacity, but it's better than
setting them on fire.

I also thought about a more complicated circuit using triacs or
something that would let 208 out of every 240 cycles pass through.

What about a 240V light dimmer? (Each panel is 375W)

Bulky, expensive transformers aren't an option, because I'll most
likely be better off buying the correct 240V panels at non-surplus
prices.

Ideas?
John

 

[Tim Shoppa]

What about a 240V light dimmer? (Each panel is 375W)

Sounds like a reasonable choice, but you have to worry about someone
cranking it all the way up. And cheap light dimmers make a lot of RFI.

Can you just run the panels from 120V (say, put two in series if from
240V)? If they really are surplus/cheap it might make sense to spread
out the panels anyway.

Tim.

 

[John Sevinsky]

If I did use a light dimmer, I would install it above the suspended
ceiling so nobody would mess with it.

Yes, I could run them off of 120V, but that would reduce these 375W
heaters to 94W if I did my math right, so I would need four times as
many. And the surplus price is more than 1/4 the new price.

John

 

[John Fields]

I've always wanted radiant ceiling panel heat in our basement, but
those panels are expensive. Recently, I found a supply of used panels
for a reasonable price, so I bought one to play with. (It's a Qmark
Aztec 375W panel.) Unfortunately, it is rated for 208V, and our house
has 240V. I hooked it up, and it worked. But, the surface temperature
in the middle of the panel reached a maximum temperature of 250F, which
might be beyond the safe operating range.

So, I'm wondering how I can reduce the power supplied to these panels.
Although this won't be perfect, I thought about some diode arrangement
that would supply power to a pair of these panels in parallel for one
half of the cycle and in series for the other half. So, they would be
getting 240V half of the time and 120V half of the time. That should
average out to less than their rated capacity, but it's better than
setting them on fire.

I also thought about a more complicated circuit using triacs or
something that would let 208 out of every 240 cycles pass through.

---
Very clever; I like it!

Easy to do, too.

Use a free-running 8 bit counter clocked by the mains, and when it
overflows set a latch which turns on a zero-crossing TRIAC driver
which turns on a TRIAC which turns on the panel. Then, when the
counter gets to 222, decode that and reset the latch. That'll turn
off the panel until the counter overflows and the cycle begins anew.

Want a schematic?

 

[Spehro Pefhany]

If I did use a light dimmer, I would install it above the suspended
ceiling so nobody would mess with it.

Yes, I could run them off of 120V, but that would reduce these 375W
heaters to 94W if I did my math right, so I would need four times as
many. And the surplus price is more than 1/4 the new price.

John

The triac or SCRs will cause RFI and will generally fail "on". They
will probably see a lot of thermal cycling if you are using a
conventional thermostat to control them.

I suggest a ~32VAC buck transformer or a(n) (about an order of
magnitude heavier) surplus stepdown transformer, from safety
considerations. Such tranformers can sometimes be had for the asking
from companies. It's part of your house-- do you really want to
compromise safety (and probably violate code)?


Best regards,
Spehro Pefhany

 

[Jim Thompson]

---
Very clever; I like it!

Easy to do, too.

Use a free-running 8 bit counter clocked by the mains, and when it
overflows set a latch which turns on a zero-crossing TRIAC driver
which turns on a TRIAC which turns on the panel. Then, when the
counter gets to 222, decode that and reset the latch. That'll turn
off the panel until the counter overflows and the cycle begins anew.

Want a schematic?

Back in my disco days I dimmed neon by cycle skipping. Caution: If a
transformer is involved pass WHOLE cycles only.

...Jim Thompson

 

[Joel Kolstad]

Back in my disco days I dimmed neon by cycle skipping. Caution: If a
transformer is involved pass WHOLE cycles only.

I'm sure we'd love to see some pictures of that on ABSE, Jim.

I had a co-worker once who claimed he and his fraternity arranged a
"headphones only" dance night... he built a huge distribution panel/amplifier,
and everyone dancing had to wear headphones plugged into the ceiling to hear
the music.

And supposedly women actually attended this shin-dig. (This would have been
in the late '70s/early '80s.)

Which I find quite surprising...

It sounds like a scene out of Revenge of the Nerds...

 

[Jim Thompson]

I'm sure we'd love to see some pictures of that on ABSE, Jim.

BC, before CAD... I MAY have pencil schematics somewhere.

But really quite simple, TRIAC's driving the transformer... pass whole
cycles, skip cycles... ratio determines dimming... worked quite well.

Around 1980, installed in Bobby McGee's on I17, along with my fabulous
boom-box.

Don't know if it's still there or not.

I had a co-worker once who claimed he and his fraternity arranged a
"headphones only" dance night... he built a huge distribution panel/amplifier,
and everyone dancing had to wear headphones plugged into the ceiling to hear
the music.

And supposedly women actually attended this shin-dig. (This would have been
in the late '70s/early '80s.)

Which I find quite surprising...

It sounds like a scene out of Revenge of the Nerds...

...Jim Thompson

 

[Jim Thompson]

Pass 7 cycles, skip the 8th

Back in my disco days I dimmed neon by cycle skipping. Caution: If a
transformer is involved pass WHOLE cycles only.

...Jim Thompson

...Jim Thompson

 

[John Sevinsky]

I'll bet that system used 1/4" jacks for the headphones.

John

 

[Homer J Simpson]

If I did use a light dimmer, I would install it above the suspended
ceiling so nobody would mess with it.

Yes, I could run them off of 120V, but that would reduce these 375W
heaters to 94W if I did my math right, so I would need four times as
many. And the surplus price is more than 1/4 the new price.

After the fire when the insurance investigator is looking at the remnants
what will your story be? Find some code approved way to run them off a split
240 V system - it might be worth a chat with the local electrical inspector.









--

 

[Paul E. Schoen]

I've always wanted radiant ceiling panel heat in our basement, but
those panels are expensive. Recently, I found a supply of used panels
for a reasonable price, so I bought one to play with. (It's a Qmark
Aztec 375W panel.) Unfortunately, it is rated for 208V, and our house
has 240V. I hooked it up, and it worked. But, the surface temperature
in the middle of the panel reached a maximum temperature of 250F, which
might be beyond the safe operating range.

It might be possible to install a small disc thermostat to the panel to
shut it off at a safe temperature.

So, I'm wondering how I can reduce the power supplied to these panels.
Although this won't be perfect, I thought about some diode arrangement
that would supply power to a pair of these panels in parallel for one
half of the cycle and in series for the other half. So, they would be
getting 240V half of the time and 120V half of the time. That should
average out to less than their rated capacity, but it's better than
setting them on fire.

I also thought about a more complicated circuit using triacs or
something that would let 208 out of every 240 cycles pass through.

What about a 240V light dimmer? (Each panel is 375W)

Bulky, expensive transformers aren't an option, because I'll most
likely be better off buying the correct 240V panels at non-surplus
prices.

You could use a small buck/boost transformer. You only need one with 240
VAC input and 24 VAC or 32 VAC output at about 1.5 A. This should be a
standard control transformer of 36 VA. The output is wired to "buck" the
input voltage, so you will get 240-24 = 216 volts. Close enough. Or 240-32
= 208 volts.

Paul

 

[John Sevinsky]

So, let's see if I got this right. One 375W 240V panel would use a
little over 1.5A. Let's say I need six for the room in our basement.
That would a little over 9A, so let's round it up to 10A. If I wanted
to use a 32V buck transformer, that would be 320VA or 0.32KVA. Is that
correct?

John

 

[Terran Melconian]

Although this won't be perfect, I thought about some diode arrangement
that would supply power to a pair of these panels in parallel for one
half of the cycle and in series for the other half. So, they would be
getting 240V half of the time and 120V half of the time. That should
average out to less than their rated capacity, but it's better than
setting them on fire.

Were you able to construct such an arrangement? I do not think it can
be done with just diodes, unfortunately.

I also thought about a more complicated circuit using triacs or

The 208v 375w panel, when hot and running at the correct voltage, would
draw 1.8A. So, how about putting 3 12v 20watt light bulbs in series
with each? The light bulbs themselves will also provide heat at nearly
perfect efficiency, and presumably the light can be productively used
somehow.

 

[Homer J Simpson]

message

The 208v 375w panel, when hot and running at the correct voltage, would
draw 1.8A. So, how about putting 3 12v 20watt light bulbs in series
with each? The light bulbs themselves will also provide heat at nearly
perfect efficiency, and presumably the light can be productively used
somehow.

There will be 240 V across these if they blow - they may not be rated for
that. Be careful with such a set-up.

 

[Joel Kolstad]

BC, before CAD... I MAY have pencil schematics somewhere.

I actually meant that we'd love to see pictures of you disco-ing.

Back in the '80s I built a phase-controlled TRIAC dimmer... no filtering, just
hard-switching of the gate, etc. The old WW-II vet running the electronics
shop told me it was a horrible design, that I'd create RFI all over the place,
just turn on the radio.

So we did (a big old tube set with a foot-wide spider web antenna).

No significant RFI.

Being a young punk I concluded he didn't know what he was talking about.

But now I know better.

Plus I now have access to spectrum analyzers!

 

[John Sevinsky]

Were you able to construct such an arrangement? I do not think it can
be done with just diodes, unfortunately.

I think I did, but my ASCII art isn't so good, so I'll to describe my
sketch.

Represent the heating elements by drawing two resistors in series
stacked vertically on the paper, making a top, middle, and bottom node.

One leg of the AC goes to a diode "pointing to" the middle node. The
same leg also goes through another diode "pointing away from" the
bottom node.

The other of the AC leg gets directly connected to the top node. The
same leg also goes to a third diode "pointing away from" the bottom
node.

I hope that's not too confusing.

John

 

[John Fields]

Back in my disco days I dimmed neon by cycle skipping.

 

[Jim Thompson]

I actually meant that we'd love to see pictures of you disco-ing.

Me ? I'm a nerd... I designed that stuff, and actually enjoyed
that variety of music, but I'm not a dancer.

Back in the '80s I built a phase-controlled TRIAC dimmer... no filtering, just
hard-switching of the gate, etc. The old WW-II vet running the electronics
shop told me it was a horrible design, that I'd create RFI all over the place,
just turn on the radio.

So we did (a big old tube set with a foot-wide spider web antenna).

No significant RFI.

Being a young punk I concluded he didn't know what he was talking about.

But now I know better.

Plus I now have access to spectrum analyzers!

...Jim Thompson

 

[John Fields]

Pass 7 cycles, skip the 8th