Muliple Peltier elements: parallel or series?

tm wrote:
> 
> "Phil Hobbs" <pcdhSpamMeSenseless@electrooptical.net> wrote in message
> news:5008B5EA.D651DCA@electrooptical.net...
> > tm wrote:
> >>
> >> "Phil Hobbs" <pcdhSpamMeSenseless@electrooptical.net> wrote in message
> >> news:500898F2.67E349F@electrooptical.net...
> >> > Don Lancaster wrote:
> >> >>
> >> >> On 7/19/2012 12:25 PM, George Herold wrote:
> >> >> > On Jul 19, 11:44 am, Syd Rumpo <use...@neonica.co.uk> wrote:
> >> >> >> On 19/07/2012 15:45, Jan Panteltje wrote:
> >> >> >>
> >> >> >>> On a sunny day (Thu, 19 Jul 2012 14:50:00 +0100) it happened
> >> >> >>> Martin
> >> >> >>> Brown
> >> >> >>> <|||newspam...@nezumi.demon.co.uk> wrote in
> >> >> >>> <1gUNr.50376$iI7.22...@newsfe03.iad>:
> >> >> >>
> >> >> >>>> The aim would be for an Xmas science demo to harvest some of the
> >> >> >>>> ~100W
> >> >> >>>> waste heat of a candle flame and drive a ~1W LED. This seemed a
> >> >> >>>> modest
> >> >> >>>> aim when I started out but in practice it proved impossible. To
> >> >> >>>> be
> >> >> >>>> any
> >> >> >>>> good the LED must be a *lot* brighter than the candle flame!
> >> >> >>
> >> >> >>> I would use a thermocouple (or some i nseries) and a JFET
> >> >> >>> oscillator
> >> >> >>> with step up transformer.
> >> >> >>> JFET oscillators oscillate from about next to zero volts up.
> >> >> >>> Have not tried it on a thermocouple, but did try the low voltage
> >> >> >>> JFET
> >> >> >>> oscillator.
> >> >> >>> Only a few parts needed...
> >> >> >>
> >> >> >> I'd go with several thermocouples in series.  The advantage over a
> >> >> >> Peltier is that you can have the cold junctions some way from the
> >> >> >> heat.
> >> >> >>
> >> >> >> Say use Iron/Constantan at about 55uV/K so for 1.8V at 200C you'd
> >> >> >> need
> >> >> >> about 160 thermocouples which is doable I reckon, probably many
> >> >> >> fewer
> >> >> >> depending on flame temperature.  You can just twist the ends
> >> >> >> together
> >> >> >> to
> >> >> >> make the junctions, should last a while.
> >> >> >>
> >> >> >> Cheers
> >> >> >> --
> >> >> >> Syd
> >> >> >
> >> >> > Hey, speaking of thermopiles I heard Hamamatsu has a new line of
> >> >> > them.
> >> >> > http://www.hamamatsu.com/news/2011/2011_09_21.html
> >> >> >
> >> >> > I can't find a spec sheet or price though.
> >> >> >
> >> >> > George H.
> >> >> >
> >> >>
> >> >> Peltier Coolers are almost always totally worthless.
> >> >> Their delta T across the heatsink usually EXCEEDS their intended delta
> >> >> T
> >> >> drop
> >> >>
> >> >> See < http://www.tinaja.com/glib/hack68.pdf >
> >> >>
> >> >> --
> >> >> Many thanks,
> >> >>
> >> >> Don Lancaster
> >> >
> >> > Nonsense.
> >> >
> >> > If you try cooling a CPU or a voltage regulator with one, you deserve
> >> > what you get.  But many kinds of instruments would be far harder to
> >> > build without them.  Oh, and of course the telecom infrastructure
> >> > wouldn't work without them either, because Peltiers are what keep all
> >> > the DWDM lasers from scribbling across neighbouring channels.
> >> >
> >>
> >> I think he implies that in his paper. Certain special applications are
> >> ideal
> >> for TECs.
> >>
> >> tm
> >
> > Setting up a straw man and then knocking it down is an old rhetorical
> > amusement, but IMO it's beneath engineers of Don's calibre.  Of course
> > he's been in the biz for probably 50 years or more, and so feels
> > entitled to take the occasional potshot from his porch, and who's to say
> > that isn't OK.
> >
> > I bought a copy of his TTL Cookbook in about 1974, when I was 14 years
> > old, and it helped me a lot back then, and for several years
> > thereafter.  I was also intrigued by his TV Typewriters, but since I
> > couldn't type they weren't that compelling.  They were really just a
> > tease anyway, aimed at those of us who didn't have our own private
> > PDP-10.
> >
> > However, saying that "Peltier coolers are almost always totally
> > worthless", when in fact they enable most of modern communications and
> > the Internet, is just plain nonsense, and needs to be called out.  Maybe
> > it'll improve Don's aim. ;)
> >
> > Cheers
> >
> > Phil Hobbs
> >
> > --
> 
> Well, saying "they enable most of modern communications and the internet" is
> a bit Algoreish in its own right. :)
> 
> I have yet to see a 10 gig SFP long haul module that is cooled. I would
> guess a good bit of the Internet uses them.
> 
> tm

I don't agree.  If you can't keep DWDM lasers tuned to within +-10 GHz
or so, you don't have the modern Internet.

Cheers

Phil Hobbs


-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

On a sunny day (Thu, 19 Jul 2012 14:31:35 -0500) it happened Jon Elson
<jmelson@wustl.edu> wrote in <HqednTtKXpNZw5XNnZ2dnUVZ_rSdnZ2d@giganews.com>:

>Martin Brown wrote:
>
>
>
>> The aim would be for an Xmas science demo to harvest some of the ~100W
>> waste heat of a candle flame and drive a ~1W LED. This seemed a modest
>> aim when I started out but in practice it proved impossible. To be any
>> good the LED must be a *lot* brighter than the candle flame!
>> 
>You are using the wrong junction for this purpose.  There are far
>better junctions made specifically for this, although the thermocouple
>in a gas furnace will almost do it.  About 3 or 4 in series will
>light an LED plenty brightly, in fact you'll need a current limiting
>resistor.  I think they use copper and iron, nothing exotic at all,
>and in a gas flame, I think you get 600 mV out of these.  A candle
>flame should be a bit less, so 4 junctions at 400 mV each should
>light a red LED.
>
>Jon

Calculating for 200 C, with my 'th' program,
I get the following voltages for the different themocouples:

Panteltje (c) th-0.5 thermocouple voltage (in mV) to temperature and reverse calculator.
Usage:
th -a [E|J|K|R|S|T] [-d] [-h] [-m C|K|F] -v voltage || -t temperature
-a type         thermocouple type, either E, J, K, R, S, or T.
-c temperature  cold junction temperature in units as described with -m.
-d              debug mode, prints functions and arguments, and some variables.
-h              help, this help.
-m mode         temperature mode for temperature to voltage, either C, K, or F for Celsius, Kelvin, or Fahrenheit.
-v voltage      voltage in mV, for voltage to temperature conversion.
-t temperature  temperature.

Type K:
panteltje: ~/compile/pantel/gpspc # th -a K -c 25 -m C -t 200
voltage is 7.138231 mV

Type E:
panteltje: ~/compile/pantel/gpspc # th -a E -c 25 -m C -t 200
voltage is 11.926184 mV

Type J:
panteltje: ~/compile/pantel/gpspc # th -a J -c 25 -m C -t 200
voltage is 9.501458 mV

Type R:
panteltje: ~/compile/pantel/gpspc # th -a R -c 25 -m C -t 200
voltage is 1.328004 mV

Type S:
panteltje: ~/compile/pantel/gpspc # th -a S -c 25 -m C -t 200
voltage is 1.298185 mV

Type T:
panteltje: ~/compile/pantel/gpspc # th -a T -c 25 -m C -t 200
voltage is 8.296125 mV

So looks like a type E gives the most
You can buy a pack of 5 of type E at omega for abut 14 UKP:
 http://www.omega.co.uk/ppt/pptsc.asp?ref=IRCO-BW

That makes 5 x 12 = 60 mV
Should run a JFET oscillator.
I have use 200 C as example, candle flame is much hotter,
these small themocouples do not interfere with the visibility of the flame I think.
I uses 200 as I solder some togetehr myself (not type E), and the solder melts at 200...

The 'th' (Linux) program is available from me.
 http://panteltje.com/panteltje/newsflex/th-0.4.tgz


Hey. Microsoft made a loss for the firt time this year,
wonder when Ba;lmer get lynched...
and it will be bought by 'investors', broken up, renamed,
burried...

Fun fun fun :-)

Jon Elson <jmelson@wustl.edu> writes:

> John Devereux wrote:
>
>
>> It was 20 years ago but IIRC...
>> 
>> In my old flat I was surprised to discover than the boiler gas valve was
>> powered directly from a thermocouple assembly in the gas flame. (I only
>> realised this after finding what I assumed was a "thermal switch" had a
>> permanent short circuit and the valve had no external power source...)
> This is called a millivolt system, still fairly common.  Uses no mains
> electricity.  It basically uses the same thermocouple used in the
> common gas safety controls on water heaters and furnaces.  The solenoid
> coil is a small number of turns of heavy-gauge copper wire, and the
> current to operate the solenoid is up to several Amps.

Pretty clever really, you can see why I was confused. I had even
designed a temperature controller with a thermocouple input by then. But
I was not used to thinking of them generating amps!

-- 

John Devereux

On 19/07/2012 20:12, Tim Wescott wrote:
> On Thu, 19 Jul 2012 14:11:44 -0500, Tim Wescott wrote:
>
>> On Thu, 19 Jul 2012 14:50:00 +0100, Martin Brown wrote:
>>
>>> The aim would be for an Xmas science demo to harvest some of the ~100W
>>> waste heat of a candle flame and drive a ~1W LED. This seemed a modest
>>> aim when I started out but in practice it proved impossible. To be any
>>> good the LED must be a *lot* brighter than the candle flame!
>>>
>>> I am guessing that to stand any chance I also need a fan assisted heat
>>> sink and to be authentic it must all be powered by the TECs. It is for
>>> a physics demo so hiding a button cell somewhere is just not on...
>>
>> You may do much better with a fan-powered Sterling engine.
>
> CANDLE-powered Sterling.  Oops.

Trouble is I had already done a candle powered heat engine and then a 
more sophisticated coffee cup powered rotary motion one in previous 
years. I was trying to find yet another candle based demo for Xmas.

-- 
Regards,
Martin Brown

On 19/07/2012 20:31, Jon Elson wrote:
> Martin Brown wrote:
>
>
>
>> The aim would be for an Xmas science demo to harvest some of the ~100W
>> waste heat of a candle flame and drive a ~1W LED. This seemed a modest
>> aim when I started out but in practice it proved impossible. To be any
>> good the LED must be a *lot* brighter than the candle flame!
>>
> You are using the wrong junction for this purpose.  There are far
> better junctions made specifically for this, although the thermocouple
> in a gas furnace will almost do it.  About 3 or 4 in series will
> light an LED plenty brightly, in fact you'll need a current limiting
> resistor.  I think they use copper and iron, nothing exotic at all,
> and in a gas flame, I think you get 600 mV out of these.  A candle
> flame should be a bit less, so 4 junctions at 400 mV each should
> light a red LED.

My back of the envelope suggests that for a candle flame delta-T of 
1000K there would be around 12mV per junction pair to play with.

I presume the boiler thermostat is a big bunch in series. A pair of 
those would provide enough power capture with a bit of luck.
How big are they?

I did try messing about with iron and copper wire too, but it didn't 
look good and produced a miniscule voltage for the amount of effort. 
Whoever said 160 reliable thermocouple wire junctions in a candle flame 
is "doable" is something of an optimist. I got fed up at about a dozen.

I didn't know about these gas boiler thermopile contraptions.

-- 
Regards,
Martin Brown

On 20/07/2012 09:05, Martin Brown wrote:

<snipped>

> I presume the boiler thermostat is a big bunch in series.

No, just one junction.  It provides enough current to *hold* a 
low-voltage solenoid valve open against a spring, but you need to 
manually hold the valve open - usually a button - until the pilot flame 
has heated the thermocouple.  This may take a minute.

I had a boiler once where the pilot kept going out and needed 
relighting.  It was an 'instant heat' boiler which only turned on the 
main burners when a hot water tap was turned on.  Eventually I got 
around to taking a look and discovered a 2" layer of dead wasps inside. 
  Seems they'd been attracted in from a nearby nest by the warmth of the 
pilot flame, then incinerated as someone turned on a tap.  Over and over 
again for many weeks, occasionally snuffing out the pilot flame on their 
way.

Cheers
-- 
Syd

On 19/07/2012 20:31, Jon Elson wrote:
> Martin Brown wrote:
>
>> The aim would be for an Xmas science demo to harvest some of the ~100W
>> waste heat of a candle flame and drive a ~1W LED. This seemed a modest
>> aim when I started out but in practice it proved impossible. To be any
>> good the LED must be a *lot* brighter than the candle flame!
>>
> You are using the wrong junction for this purpose.  There are far
> better junctions made specifically for this, although the thermocouple
> in a gas furnace will almost do it.  About 3 or 4 in series will
> light an LED plenty brightly, in fact you'll need a current limiting
> resistor.  I think they use copper and iron, nothing exotic at all,
> and in a gas flame, I think you get 600 mV out of these.  A candle
> flame should be a bit less, so 4 junctions at 400 mV each should
> light a red LED.

Seems to be entirely wrong order of magnitude here - at least in the UK 
these gas thermocouples seem to provide about 30mV in a flame. Though 
apparently at sufficient current to pull in some kind of solenoid.

http://ecc.emea.honeywell.com/downloads/MU1R9124.PDF

A pair or more would generate enough voltage to run a step up. This 
might be more satisfactory for demo than flat plates over the flame.

-- 
Regards,
Martin Brown

On Fri, 20 Jul 2012 08:45:18 +0100, Martin Brown wrote:

> On 19/07/2012 20:12, Tim Wescott wrote:
>> On Thu, 19 Jul 2012 14:11:44 -0500, Tim Wescott wrote:
>>
>>> On Thu, 19 Jul 2012 14:50:00 +0100, Martin Brown wrote:
>>>
>>>> The aim would be for an Xmas science demo to harvest some of the
>>>> ~100W waste heat of a candle flame and drive a ~1W LED. This seemed a
>>>> modest aim when I started out but in practice it proved impossible.
>>>> To be any good the LED must be a *lot* brighter than the candle
>>>> flame!
>>>>
>>>> I am guessing that to stand any chance I also need a fan assisted
>>>> heat sink and to be authentic it must all be powered by the TECs. It
>>>> is for a physics demo so hiding a button cell somewhere is just not
>>>> on...
>>>
>>> You may do much better with a fan-powered Sterling engine.
>>
>> CANDLE-powered Sterling.  Oops.
> 
> Trouble is I had already done a candle powered heat engine and then a
> more sophisticated coffee cup powered rotary motion one in previous
> years. I was trying to find yet another candle based demo for Xmas.

Well, in that case, make a padded bear-trap gizmo that's triggered by 
someone sticking their finger into a hole.  Connect it to a string that's 
connected to a generator that charges up a capacitor, with the capacitor 
powering some LEDs.  Set the bear trap up so that after the string is 
pulled out a foot or so, it releases.

Set up the thing so that once the finger in question triggers the bear 
trap, a candle flame is brought to bear directly on the finger.

Voila!  An LED lighter that's powered by candle flame and gullibility.  
Not only does it have the advantage of being unique, but you won't have 
to worry about thinking of something next year, because you won't be 
invited back.

-- 
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Spehro Pefhany wrote:


> 
> Those must be thermopiles, a single Cu-Fe thermocouple would produce
> MUCH less voltage.
Well, I've only picked at a BAD one, and it was quite burned up after
years of exposure to the pilot flame.  But, it couldn't have had more
than a couple of junctions in a coaxial arrangement.  I tend to think it
really was just a single junction.  I'll try to find details.  But, maybe
the voltage produced is way lower than what I suggested.

Jon

On 20 Jul., 01:49, "k...@att.bizzzzzzzzzzzz" <k...@att.bizzzzzzzzzzzz>
wrote:
> On Thu, 19 Jul 2012 16:38:19 -0700 (PDT), "langw...@fonz.dk"
>
>
>
>
>
>
>
>
>
> <langw...@fonz.dk> wrote:
> >On 20 Jul., 00:59, "k...@att.bizzzzzzzzzzzz" <k...@att.bizzzzzzzzzzzz>
> >wrote:
> >> On Thu, 19 Jul 2012 15:14:04 -0400, Phil Hobbs
>
> >> <pcdhSpamMeSensel...@electrooptical.net> wrote:
> >> >Tim Wescott wrote:
>
> >> >> On Thu, 19 Jul 2012 14:11:44 -0500, Tim Wescott wrote:
>
> >> >> > On Thu, 19 Jul 2012 14:50:00 +0100, Martin Brown wrote:
>
> >> >> >> On 19/07/2012 14:29, Phil Hobbs wrote:
> >> >> >>> Martin Brown wrote:
>
> >> >> >>>> Current. They are in essence a bunch of high power diode PN
> >> >> >>>> junctions.
>
> >> >> >>> I'd control them individually, or wire them in series. =A0Compo=
und
> >> >> >>> Peltiers have a nasty thermal instability, where they start get=
ting
> >> >> >>> hotter as you go to higher current instead of colder.
>
> >> >> >>> If you put them in series, they all see the same current, which=
 along
> >> >> >>> with a spreader plate top and bottom) helps prevent runaway. =
=A0If you
> >> >> >>> put them in parallel, the ones with higher delta-T will draw le=
ss
> >> >> >>> current than the lower delta-T ones, which gives you a nice
> >> >> >>> stabilizing action at low drive current.
>
> >> >> >>> If you push them, or you lose cooling water to your heat exchan=
ger,
> >> >> >>> the sign of the gain can invert and the whole thing turns to la=
va very
> >> >> >>> fast. =A0(That can happen with individual control or series-con=
nection
> >> >> >>> as well, but the melty ones can't hog current from the cooler o=
nes, so
> >> >> >>> it isn't so unstable.)
>
> >> >> >>> The other thing is that you have to allow the Peltiers to slide=
 around
> >> >> >>> a bit as the cold plate cools down, because otherwise they'll c=
rack.
> >> >> >>> So use Arctic Silver or a very small amount of very good therma=
l
> >> >> >>> grease on one side, and solder on the other. =A0IIRC you can so=
lder the
> >> >> >>> 30 mm ones, but not the 50 mm ones--it'll be in the datasheet. =
(Solder
> >> >> >>> is about 100x better than ordinary thermal grease.)
>
> >> >> >>> I talk a lot about Peltiers in my free thermal chapter,
> >> >> >>>http://electrooptical.net/www/book/thermal.pdf. =A0There's also =
an
> >> >> >>> incomplete draft that was intended to accompany the second edit=
ion,
> >> >> >>> which has some more stuff in it, at
> >> >> >>>http://electrooptical.net/www/book/draftthermal.pdf.
>
> >> >> >> Handy info thanks. Although the TECs are available their datashe=
ets are
> >> >> >> often inadequate and miss out details needed for real world desi=
gns.
>
> >> >> >> Can I pick your brains for a slightly odd requirement? Ideally o=
ne that
> >> >> >> can be done with at most a pair of TECs one high temp and one no=
rmal.
> >> >> >> On paper they were rated for 3A and 14v or so each but in practi=
ce I
> >> >> >> was struggling in reverse to get 2v and a few mA out with a cand=
le
> >> >> >> flame heated plate at 250C on one side and a slab of aluminium a=
t -18C
> >> >> >> on the other. I gave up in the end as it was taking too long and=
 safety
> >> >> >> concerns of very hot metal plates and children scuppered it in t=
he end.
>
> >> >> >> The aim would be for an Xmas science demo to harvest some of the=
 ~100W
> >> >> >> waste heat of a candle flame and drive a ~1W LED. This seemed a =
modest
> >> >> >> aim when I started out but in practice it proved impossible. To =
be any
> >> >> >> good the LED must be a *lot* brighter than the candle flame!
>
> >> >> >> I am guessing that to stand any chance I also need a fan assiste=
d heat
> >> >> >> sink and to be authentic it must all be powered by the TECs. It =
is for
> >> >> >> a physics demo so hiding a button cell somewhere is just not on.=
..
>
> >> >> > You may do much better with a fan-powered Sterling engine.
>
> >> >> CANDLE-powered Sterling. =A0Oops.
>
> >> >> > Or, for a complete demo, have one TE element complete with lots o=
f hype,
> >> >> > and one "150 year old technology" Sterling engine for comparison.=
..
>
> >> >> --
>
> >> >Of course, TECs are 60-year-old technology....
>
> >> Yup. =A0My father designed a TEC refrigerator in the late '50s or very=
 early
> >> '60s, complete with a solid-state power supply (string of Germanium DO=
-3
> >> PNPs). =A0I had a couple of them in the basement to play with when I w=
as a kid.
> >> It was designed to go in a car but never made it past prototype. =A0It=
 got beat
> >> out by Styrofoam disposables. =A0;-)
>
> >you can get plenty of 12V TEC refrigerators/coolers but I don't think
> >they have
> >much if any smarts in them, they just say something like "up to 20'C
> >below ambient"
>
> Fifty years ago you couldn't. =A0;-)

my dad was still in school then so I wouldn't know :p

-Lasse