Searching \ for '[EE:] About test boxes (hot and cold)' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=about+test+boxes
Search entire site for: 'About test boxes (hot and cold)'.

Exact match. Not showing close matches.
PICList Thread
'[EE:] About test boxes (hot and cold)'
2004\01\16@050302 by 8859-9?B?1m1lciBZYWxo/Q==?=

I don't know if this is EE or OT; however, I need a box where I can set the
temprature inside it (within a certain error margin) and keep it the same
temprature (again within a certain error margin).  I will use this box to
put some electronic boards into temprature tests.

Is there a name for this "box"?  Any recommendations?

Thanks in advance,

--
http://www.piclist.com hint: To leave the PICList
spam_OUTpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu

2004\01\16@050511 by Hulatt, Jon

flavicon
face
An oven?

{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspam_OUTspamTakeThisOuTmitvma.mit.edu

2004\01\16@052003 by Russell McMahon

face
flavicon
face
> I don't know if this is EE or OT; however, I need a box where I can set
the
> temprature inside it (within a certain error margin) and keep it the same
> temprature (again within a certain error margin).  I will use this box to
> put some electronic boards into temprature tests.
>
> Is there a name for this "box"?  Any recommendations?

EE I'd say.

One name would be an "environmental chamber". Depending on how hot or how
cold you want to go, a Peltier effect controlled refrigerator as sold in
many US stores may do the trick. The Peltier device can either heat or cool
(by reversing polarity). Typical cool boxes have a 50 watt Peltier and a fan
and are large enough to take 4 to 6 large soft drink bottles.

A controller can be made that will seamlessly switch the Peltier device from
heating to cooling as required.



       RM


___________________________________________________________
If we have already accidentally created an unexpectedly lethal virus
by single gene transfer, why should we believe that it might not happen
again, once or on numerous occasions in future, but in less controlled
circumstances?

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamspam_OUTmitvma.mit.edu

2004\01\16@054116 by 8859-9?B?1m1lciBZYWxo/Q==?=

Russel thank you for the answer.  I will do a search for Peltier and
environmental chamber and see what comes up and go from there.  I need
something that I can set between -25°C to +75°C (-13°F to +167°F) plus or
minus 5°C, I don't need a very precise instrument.

Thank you,


{Original Message removed}

2004\01\16@061543 by hael Rigby-Jones

picon face
>-----Original Message-----
>From: Ömer Yalhi [@spam@oyalhiKILLspamspamTEKSAN.COM.TR]
>Sent: 16 January 2004 10:44
>To: KILLspamPICLISTKILLspamspamMITVMA.MIT.EDU
>Subject: Re: [EE:] About test boxes (hot and cold)
>
>
>Russel thank you for the answer.  I will do a search for
>Peltier and environmental chamber and see what comes up and go
>from there.  I need something that I can set between -25°C to
>+75°C (-13°F to +167°F) plus or minus 5°C, I don't need a very
>precise instrument.
>
>Thank you,

I doubt very much if a cheap food cooler will be able to get anywhere near
the temperatures you want, especialy the lower range.  If the products that
you are testing at these low temperatures dissipate much power, then using
thermoelectric coolers will be difficult.  Liquid cooling the hot side of
the cooler is almost certainly going to be needed.

A proper environmental chamber is what you require, unfortunately they are
large, noisy, power hungry and most of all expensive.  However, there are
businesses that rent out time on their chambers, you just turn up with your
product and test equipment.

Regards

Mike




=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================
Any questions about Bookham's E-Mail service should be directed to
RemoveMEpostmasterTakeThisOuTspambookham.com.

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestspamBeGonespammitvma.mit.edu

2004\01\16@061958 by William Chops Westfield

face picon face
On Friday, Jan 16, 2004, at 02:43 US/Pacific, Ömer Yalhý wrote:

> Russel thank you for the answer.  I will do a search for Peltier and
> environmental chamber and see what comes up and go from there.  I need
> something that I can set between -25°C to +75°C (-13°F to +167°F) plus
> or
> minus 5°C, I don't need a very precise instrument.

the January Circuit Cellar Ink magazine had a project where the author
had converted a cheap (relatively speaking) peltier drink cooler into
a CPU-controlled temperature chamber.  It was quite a reasonable article.
I believe they only got 8 to 60C out of it, though.  Peltiers are not
very efficient coolers, so it's hard to get very low temps, especially
if whatever you're cooling is generating any heat...

Used a PIC, too, I think.

BillW

--
http://www.piclist.com hint: To leave the PICList
TakeThisOuTpiclist-unsubscribe-requestEraseMEspamspam_OUTmitvma.mit.edu

2004\01\16@063033 by hael Rigby-Jones

picon face
{Quote hidden}

Unfortunately the author decided to use thermostatic control of the TEC,
which is definitely not recommended.  The constant thermal cycling of the
TEC will cause premature failure.

A switching or linear controller is prefered, and coupled with a decent PID
controller will give improved temperature stability over a simple
thermostat.

Regards

Mike




=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================
Any questions about Bookham's E-Mail service should be directed to
EraseMEpostmasterspambookham.com.

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestEraseMEspamEraseMEmitvma.mit.edu

2004\01\16@064110 by Russell McMahon

face
flavicon
face
> Russel thank you for the answer.  I will do a search for Peltier and
> environmental chamber and see what comes up and go from there.  I need
> something that I can set between -25°C to +75°C (-13°F to +167°F) plus
> or
> minus 5°C, I don't need a very precise instrument.

That's liable to be rather too low for a simple Peltier cooler. You can get
multi stage units but cooling capacity drops and price rises alarmingly.

If I was wanting to do this "on the cheap" I'd consider using eg dry ice as
a cooling medium. You could blow air through a dry ice chamber at a variable
rate to achieve a variable rate stream of cold air and use a simple
electronic temperature sensor to control fan speed. A few polystyrene
"chilly bins" and some polystyrene insulation should suffice. On the heating
side you are unlikely to have much trouble. Light bulbs or power resistors
of some sort should suffice as heaters for small boxes. Or a fan heater.



       Russell McMahon

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestspam_OUTspamKILLspammitvma.mit.edu

2004\01\16@064524 by Jinx

face picon face
> January Circuit Cellar Ink magazine had a project

> I believe they only got 8 to 60C out of it, though.  Peltiers are not
> very efficient coolers, so it's hard to get very low temps

Perhaps a little dry ice (-79C) ?

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestTakeThisOuTspamspammitvma.mit.edu

2004\01\16@065145 by 8859-9?B?1m1lciBZYWxo/Q==?=

> A proper environmental chamber is what you require, unfortunately they are
large, noisy, power hungry and most of all expensive.  However, there are
businesses that rent out time on their chambers, you just turn up with your
product and test equipment.

I have searched the internet and found wonderful chambers; however, the cost
starts from about 5000 USD (us dollars).  I found out that they rent it as
well.  Now, I need to find a place in my country that rents them.  I doubt
it though.

It seems that I'll start my tests with a no-frost deep-freezer that my
fridge has (~18°C) and for the heat, I'll use my wifes blow dryer.  And go
from there...

Thanks everyone for their reply,

Omer

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspamspamspamBeGonemitvma.mit.edu

2004\01\16@065353 by Jake Anderson

flavicon
face
a peltier will shoot for roughly 65C temperature difference hot side to cold
side.

A peltier that pumps 72W of heat will also produce 72W of heat (IE 144 watts
must be disipated from the hot side)

you can stack them if you want
IE 1 peltier level 1 then 2 at level 2 etc

They do < 0 temps quite easily.
There are some issues with switching them I think you want it > 2khz or
really slow but i dunno that for sure.
if you are in australia oatley electronics sells them pretty cheap.

> {Original Message removed}

2004\01\16@070354 by 8859-9?B?1m1lciBZYWxo/Q==?=

What is a dry ice?

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestKILLspamspammitvma.mit.edu

2004\01\16@071224 by Russell McMahon

face
flavicon
face
> What is a dry ice?

"Dry Ice" is solid Carbon Dioxide. Often used for cooling perishable
foodstuffs. Available commonly in many countries. Are you in Turkey? May
well be available. it is produced as a by product of gas liquefication for
medical or commercial use (eg welding).


       RM

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestSTOPspamspamspam_OUTmitvma.mit.edu

2004\01\16@071758 by Jinx

face picon face
> What is a dry ice?

Solid carbon dioxide. Quite cheap and readily available

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestSTOPspamspamEraseMEmitvma.mit.edu

2004\01\16@071802 by hael Rigby-Jones

picon face
>-----Original Message-----
>From: Jake Anderson [KILLspamgrooveeespamBeGonespamOPTUSHOME.COM.AU]
>Sent: 16 January 2004 12:53
>To: EraseMEPICLISTspamEraseMEMITVMA.MIT.EDU
>Subject: Re: [EE:] About test boxes (hot and cold)
>
>
>a peltier will shoot for roughly 65C temperature difference
>hot side to cold side.
>

As an aside,  many people don't realise that the heat pumping capacity of
the TEC falls as the temperature difference across the TEC increases.  As
the temperature difference approaches the maximum for any particular device,
the heat pumping capacity drops to zero. The power rating given to TEC's is
for a temperature difference of zero.

Regards

Mike




=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================
Any questions about Bookham's E-Mail service should be directed to
@spam@postmaster@spam@spamspam_OUTbookham.com.

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestspamKILLspammitvma.mit.edu

2004\01\16@073936 by 8859-9?B?1m1lciBZYWxo/Q==?=

I'll ask my wife, she is a pharmacist maybe she has it in her store?  I'll
let you know.

-----Original Message-----
From: pic microcontroller discussion list [.....PICLISTspam_OUTspamMITVMA.MIT.EDU] On
Behalf Of Russell McMahon
Sent: Friday, January 16, 2004 2:11 PM
To: TakeThisOuTPICLIST.....spamTakeThisOuTMITVMA.MIT.EDU
Subject: Re: [EE:] About test boxes (hot and cold)

> What is a dry ice?

"Dry Ice" is solid Carbon Dioxide. Often used for cooling perishable
foodstuffs. Available commonly in many countries. Are you in Turkey? May
well be available. it is produced as a by product of gas liquefication for
medical or commercial use (eg welding).


       RM

--
http://www.piclist.com hint: To leave the PICList
TakeThisOuTpiclist-unsubscribe-requestKILLspamspamspammitvma.mit.edu

2004\01\16@091707 by Olin Lathrop

face picon face
Vmer Yalhi wrote:
> What is a dry ice?

Frozen carbon dioxide.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestspamRemoveMEmitvma.mit.edu

2004\01\16@092923 by Paul Hutchinson

picon face
They are usually called temperature chambers or, environmental chambers,
Test Equity sells a broad range of them @
http://www.testequity.com/

There are many other suppliers but, the price from any of them may be more
than you expect (~$5K to $20K new, ~$2K and up used). Depending on the
standards you must meet, temperature range and accuracy required, you may be
able to design and build your own for less money.

A google search for "temperature chamber" finds that the January edition of
Circuit Cellar magazine  http://www.circuitcellar.com/magazine/ has a build
your own temperature chamber article using a PIC16F876 that may help.

Paul

> {Original Message removed}

2004\01\16@093958 by Olin Lathrop

face picon face
Michael Rigby-Jones wrote:
> Unfortunately the author decided to use thermostatic control of the
> TEC, which is definitely not recommended.  The constant thermal
> cycling of the TEC will cause premature failure.

It's also less efficient.  The efficiency of a peltier cooler decreases with
increasing drive current.  You therefore want to use a linear drive with
these devices, probably contolled by a PID loop.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestspamspamBeGonemitvma.mit.edu

2004\01\16@095001 by Eisermann, Phil [Ridg/CO]

flavicon
face
pic microcontroller discussion list wrote:
> It seems that I'll start my tests with a no-frost deep-freezer that my
> fridge has (~18°C) and for the heat, I'll use my wifes blow
> dryer.  And go
> from there...
>
I'll offer another suggestion. The home brewing industry has these
little temperature controllers for around $150 or so. You turn your
freezer to the coldest temperature, but the freezer is plugged into
the temperature controller. The controller turns the freezer on or
off as required. It can be set for heating or cooling IIRC. So, there's the cooling side already taken care of.

For heating, you could dismantle a toaster oven and mount the coils
on a suitable bracket to the inside of the freezer so the plastic
doesn't melt. Unplug the freezer, and plug in the oven coils
instead.
It should also be simple to build a PIC circuit that could control
both heating/cooling at the same time; using ready-made components
would allow you to quickly test the feasability of this concept.

Phil Eisermann
Electronics Engineer
The Ridge Tool Company
(440)329-4680

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-request@spam@spamspam_OUTmitvma.mit.edu

2004\01\16@100458 by Edward Gisske

flavicon
face
Ah, homebrew!

I use Johnson Controls A419-ABC digital temp controls for my kegerator and
my laagering cooler (both old chest freezers). It costs about $50 from a
local HVAC supplier (G.A. Larson, who has branches around the country). It
runs on and switches 120V, has offset, anti-short-cycle delay, and
programmable dead-band. The control can be setup to turn on or off on
temperature rise. I can hold temp to within a degree or so after fine tuning
the controller.

Now go have a beer.....

Edward Gisske, P.E.
Gisske Engineering
608-523-1900
TakeThisOuTgisskespamspamoffex.com

{Original Message removed}

2004\01\16@105405 by Oyvind Tjervaag

flavicon
face
On Fri, 2004-01-16 at 13:53, Jake Anderson wrote:
> a peltier will shoot for roughly 65C temperature difference hot side to cold
> side.
>
> A peltier that pumps 72W of heat will also produce 72W of heat (IE 144 watts
> must be disipated from the hot side)

Actually I believe (or I might be very wrong...) a peltier will produce
more power than it is pumping from the cold side due to heat also being
produced internally in the plate.

I did a project once where I cooled a metal plate to about -25C it took
3 stacked peltiers to get that low. The most tricky part was to actually
'tune' the three plates (in terms of current and voltage) to get them to
work the best. I only used air cooling at the hot side for this
application. Using water would probably lower that temerature
significantly. Remember to get silicon sealed units if used for cooling.
Water could kill them quickly...

One place tor more information: http://www.melcor.com/handbook.htm

Xyvind

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestEraseMEspammitvma.mit.edu

2004\01\16@111930 by Tom

flavicon
face
By now, others have told you the name: environmental oven(or chamber).

The oven you need depends on:
a) volume of boards to be tested (cubic meters, inches, etc.)
b) temperature range - how cold to how high
c) required rate of change of temperature
d) amount of electrical power devices under test require

Chambers are available that use liquid nitrogen to go very cold very
quickly. Very $$$. You will have to decide what you need here.  If your
needs are somewhat modest, you can build your own.

My company has a need to test boards from 0 C to about 50 C.  We purchased
an upright freezer from Sears (about $250).  The heater is provided by 200
to 300 watts of lightbulbs.  The controller is an a/d io card (National
Instruments or clone) stuck in an old 486 pc.

Here is a secret on how to make it work.

Don't attempt to regulate the freezer.  The compressor won't like it. Just
turn it on and leave it on.  Regulate the temperature by adding heat.
Also, open the back of the freezer and pull out the temperature sensor
where it's attached to the freezer compartment. Just bend it out into the
air so it never tries to cycle the freezer.  Add a small fan to stir the air.

Setup your temperature sweep to go cold and ramp up to your hot
temperature. (btw, don't try to go much above 50 C; you may melt the
plastic liner of the freezer).  The pc should be setup to not only control
the temp but cycle power to the test boards and log data for you.  You will
need to know how to write a simple pid control loop program.

Run down to at least 5 degrees colder than the bottom of your test. Soak
for some period of time. Then turn on power to your boards. Now begin
ramping the temp up at your required rate.  Once you get up near room
temperature, you can turn off the freezer and continue on regulating heat
with just the heater (light bulb).

We have such a setup as described running and have used it for around 10
years.  It works perfectly for our needs.  As usual, ymmv.

Good luck!
Tom


At 12:04 PM 1/16/2004 +0200, you wrote:
>I don't know if this is EE or OT; however, I need a box where I can set the
>temprature inside it (within a certain error margin) and keep it the same
>temprature (again within a certain error margin).  I will use this box to
>put some electronic boards into temprature tests.
>
>Is there a name for this "box"?  Any recommendations?

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestEraseMEspamspam_OUTmitvma.mit.edu

2004\01\16@115630 by 8859-9?B?1m1lciBZYWxo/Q==?=

I would like to thank everone for their input on this topic.  The
environmental chamber will be too high for my purposes right now.  However,
I have got some great ideas to make my own chamber, as several and to the
point ideas are given.

I will try to make one since this is a nice product to have, it will be
usefull in the feature as well.

Thank you and best regards to all,

Omer

--
http://www.piclist.com hint: To leave the PICList
@spam@piclist-unsubscribe-requestRemoveMEspamEraseMEmitvma.mit.edu

2004\01\16@163148 by John N. Power

flavicon
face
> From:         William Chops Westfield[SMTP:EraseMEwestfwspam@spam@MAC.COM]
> Sent:         Friday, January 16, 2004 6:19 AM
> To:   @spam@PICLISTspam_OUTspam.....MITVMA.MIT.EDU
> Subject:      Re: [EE:] About test boxes (hot and cold)

> the January Circuit Cellar Ink magazine had a project where the author
> had converted a cheap (relatively speaking) peltier drink cooler into
> a CPU-controlled temperature chamber.  It was quite a reasonable
> article.
> I believe they only got 8 to 60C out of it, though.  Peltiers are not
> very efficient coolers, so it's hard to get very low temps, especially
> if whatever you're cooling is generating any heat...

> Used a PIC, too, I think.

Yes, a 16F876.

John Power

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestEraseMEspammitvma.mit.edu

2004\01\17@060819 by dr. Imre Bartfai

flavicon
face
On Fri, 16 Jan 2004, Olin Lathrop wrote:

> Vmer Yalhi wrote:
> > What is a dry ice?
>
> Frozen carbon dioxide.
>
Be extremely careful when handling as it is -72C cold and may cause severe
burn-like injuries. Nevertheless, the evaporated carbon-dioxyde would
cause suffocation when the ventillation is insufficient.

Regards,
Imre

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2004\01\17@071531 by Jake Anderson

flavicon
face
that's what I said
a single Peltier running using 72W of electricity (common size) will
dissipate at a maximum 144 watts of heat from the hot side. 72 watts from
the power in then 72 watts pumped from cold to hot side, as far as I am
aware.

so if you want to stack them and get optimum cooling you need to have them
pyramid style. 1x on the first layer, 2x on the 2nd, 4x on the 3rd etc etc.
the thermal design for such a setup is left as an exercise for the reader.


> {Original Message removed}

2004\01\17@074644 by Oyvind Tjervaag

flavicon
face
Oops.. Then I misread your mail..

Xyvind

On Sat, 2004-01-17 at 14:16, Jake Anderson wrote:
{Quote hidden}

> > {Original Message removed}

2004\01\17@113217 by Spehro Pefhany

picon face
At 12:12 PM 1/17/2004 +0100, you wrote:
>On Fri, 16 Jan 2004, Olin Lathrop wrote:
>
> > Vmer Yalhi wrote:
> > > What is a dry ice?
> >
> > Frozen carbon dioxide.
> >
>Be extremely careful when handling as it is -72C cold and may cause severe
>burn-like injuries.

Not that much colder than outside the last few days. ;-)

You can hold a chunk in your hands if you flip it back and forth, but
gloves are much better and will prevent even slight frostbite "burns" on
the skin.

>  Nevertheless, the evaporated carbon-dioxyde would
>cause suffocation when the ventillation is insufficient.

Good point. It's really pretty innocuous stuff. My only problem with it
is that I have to drive 20 min each way and buy a big sack of it for C$25
minimum from a commercial supplier. I understand some supermarkets in the
US carry it.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speffspamBeGonespaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2004\01\18@101851 by Ian McLean

flavicon
face
Olin I can confirm what you are saying from experimentation.  TEC's perform
poorly with thermostatic control.  They do not respond well to simple
current change to control them.  Apart from the inefficiency issues, one
other point is that unlike most heating and cooling systems that use
thermostats to control them, TEC's change tempurature too quickly for
thermostats to keep up with them.

However, using a well designed PID loop, running PWM at 5kHz on an F877, I
have obtained constant tempuratures between -30 and -25 degrees celcius in a
test chamber, using a single 8W TEC available from Oatley Electronics here
in Australia.

You should not try PWM frequencies below about 2kHz because it causes
thermal stress to the TEC.

I have brought this topic up before and generating a lot of criticism and
alternatives for driving TEC's.  Some will say that PWM is also not the best
way to control them.

I am not trying to rekindle that argument, just pointing out what I have
observed through a lot of empirical experimentation.

A few points to watch to get good performance from Peltier TEC's:
1) Large thermal mass, and good cooling for hot junction is essential.
2) Watch out for condensation !
3) Stacking Peltiers may be OK if you want lower cold junction tempuratures,
but the "cooling power" diminishes, so if you have any large mass to cool,
doing this generally does not help (i.e. no such thing as a free lunch).
Remember too that the hot junction of the top Peltier in a stack is going to
need a lot of cooling.  Better to use multiple Peltiers on multiple
locations of the mass to cool.  In a cool box for example, have a TEC on
each wall of the chamber, rather than stacking them.

Rgs
Ian

> {Original Message removed}

2004\01\18@111704 by Olin Lathrop

face picon face
Ian McLean wrote:
> I have brought this topic up before and generating a lot of criticism
> and alternatives for driving TEC's.  Some will say that PWM is also
> not the best way to control them.

It isn't, unless the individual pulses are filtered so that the Peltier
device sees a constant current.  This is again due to their inefficiency at
high drive current.  This means you will get more cooling at a constant half
current than alternating between off and full on half the time.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2004\01\19@030653 by hael Rigby-Jones

picon face
>-----Original Message-----
>From: Ian McLean [RemoveMEianmm@spam@spamspamBeGoneOPTUSNET.COM.AU]

>I have brought this topic up before and generating a lot of
>criticism and alternatives for driving TEC's.  Some will say
>that PWM is also not the best way to control them.
>

PWM is fine for driving TEC's, as long as it is smoothed with an LC filter.
Ripple on the TEC supply dpoes degrade efficiency.  I guess if efficiency is
not a prime concern then a high frequency unfiltered PWM would do the job.

BTW, several companies now make PWM controllers specifically designed for
driving TEC's.

Analog Devices
http://www.analog.com/UploadedFiles/Data_Sheets/35223230285638ADN8830_c.pdf

TI
focus.ti.com/docs/search/paramsearch.jhtml?familyId=578&tfsection=par
am_table&templateId=3

Maxim
http://search.maxim-ic.com/search/?goback2=http://www.maxim-ic.com/products.
cfm&sp-q=TEC&sp-a=00040d7a-sp10026bd5&sp-k=all_collection&sp-p=all

>I am not trying to rekindle that argument, just pointing out
>what I have observed through a lot of empirical experimentation.
>
>A few points to watch to get good performance from Peltier TEC's:
>1) Large thermal mass, and good cooling for hot junction is essential.

Only if you are using the TEC for cooling!

{Quote hidden}

TEC's in parallel do not increase the maximum delta T however, only the heat
pumping capacity.  Stacked Peltier modules are available commercialy.  As
mentioned by someone else, the TEC elements are stacked in order of
increasing power (and therefore size) in order to handle the heat generated
by the previous elements.

Regards

Mike




=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================
Any questions about Bookham's E-Mail service should be directed to
.....postmaster@spam@spamEraseMEbookham.com.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\01\19@074758 by Ian McLean

flavicon
face
Olin, Michael, and others experienced in this, could I press you a little
further on this issue ?

As Michael pointed out, an LC filter on the PWM output to the Peltier
smooths the current and improves efficiency.  As I said before, I have
achieved minus 30 degrees in a test box using this exact setup.  Probably
should have mentioned an LC filter, thanks for pointing that out.  If I
recall, you pointed this out to me quite a long time ago when I first
started controlling Peltier's with the PIC16F877 PWM module & FET switching.

However, when efficiency is not so much an issue, could an LC actually
create more problems than it solves ?  An example would be ringing, which
can be a real problem in a PIC environment.  What is the trade off here
between efficiency and noise ?  I have used a simple LC.  I cannot recall
right now what values I used for L and C, but I do remember just wrapping
thick magnet wire around a small ferrite toroid and using a simple mono cap,
but maybe you could suggest a better LC configuration with ideal values to
drive a 12V/8A Peltier at 5kHz ?  I have found that this is not the easiest
thing to simulate with Spice.

Rgs
Ian

> {Original Message removed}

2004\01\19@083232 by hael Rigby-Jones

picon face
{Quote hidden}

The efficiency drop is related to ripple current.  An approximation for the
typicaly efficiency loss is given by:

DT / DTmax = 1 / (1+N^2)

Where DT is the acheivable temperature delta with the applied ripple
Dtmax is the maximum temperature delta with no ripple
N is the ripple factor (i.e. ripple current/mean current)

For lower values of ripple, the performance hit is small, but climbs rapidy
due to the square term.  Most manufacturers recommend 10% or less ripple to
achieve good performance.

As for the LC filter, choose values that have a resonant frequency well
below (5-10 times less as a rule of thumb) the switching frequency ( f= 1 /
2*PI*SQR(LC) ).  Your choice of values will primarily be determined by the
size of the inductor you can accommodate, bearing in mind the current
requirements of the TEC.  5KHz is a fairly low frequency, which means large
L's and C's will be required for effective filtering.

HTH

Mike






=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================
Any questions about Bookham's E-Mail service should be directed to
postmasterEraseMEspam@spam@bookham.com.

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\01\19@084101 by Olin Lathrop

face picon face
Ian McLean wrote:
> However, when efficiency is not so much an issue, could an LC actually
> create more problems than it solves ?  An example would be ringing,
> which can be a real problem in a PIC environment.

The assumption is that the Q of the filter is rather low, and that its
resonant frequency is well below the PWM frequency.  In that case ringing is
a non-issue.

> What is the trade
> off here between efficiency and noise ?

By "noise", I guess you mean anything that isn't the desired flat drive
signal to the Peltier junction.  As long as the ripples are small compared
to the size of the drive signal, efficiency loss will also be small.

Think of a Peltier cooler as a perfect cooler in series with a resistor.
The gotcha is that the heat generated by this resistor also heats the
Peltier junctions.  The Peltier cooling effect is proportional to the
current thru the junctions, but the heat produced by the resistor is
proportional to the square of the current.  Obviously the heat from the
resistor is going to win at some point as the current is increased.  After
that point, increasing current will produce less cooling, not more.  Another
way of putting this is that the efficiency decreases with the current drive.
That's why 1A and 0A half the time is less efficient than 1/2A all the time.
Any deviation from the flat 1/2A drive current (the noise) will decrease the
overall efficiency.

> I have used a simple LC.  I
> cannot recall right now what values I used for L and C, but I do
> remember just wrapping thick magnet wire around a small ferrite
> toroid and using a simple mono cap, but maybe you could suggest a
> better LC configuration with ideal values to drive a 12V/8A Peltier
> at 5kHz ?

There is nothing wrong with wrapping some magnet wire around a steel bolt or
whatever, as long as it can take the max current without getting too hot,
the magnetic medium doesn't saturate, and you can tolerate whatever series
resistance it has.  Look at the result on a scope, and if it looks good
enough it is.

If you want to specify an off the shelf product, then you have to do some
math.  Peltier voltages are usually low, a few 10s of volts max.  Therefore
it will be cheaper and easier to get a large cap than a large inductor to
get the same low cutoff frequency.  To get a sanity check on readily
available inductors, take a look at the CoilCraft PCV series.  The largest
one that can handle 10A is 270uH.  You were using 5KHz PWM, so lets say you
want a the LC filter corner frequency to be 500Hz.  That requires a 375uF
cap.  Not bad.  A 1mF cap should be easily attainable, so you could get away
with a 100uH inductor.  Increasing the PWM frequency would allow an even
smaller inductor.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.

2004\01\19@181237 by Ian McLean

flavicon
face
Thanks to both Olin and Michael.

A combination of 100uH and 1mF seems good to me for an "off-the-shelf"
solution, without making the inductor too big.  I cannot increase the PWM
frequency much above 5kHz at this stage in the design I am using it in,
because the PIC is only running at 4MHz.  If I increase the crystal speed to
20MHz, then I could easily accomodate a 10kHz PWM frequency.  Unfortunately,
increasing crystal speed will require a considerable amount of code revision
to adjust timing, examples being PWM registers, serial registers,
bit-bashing clocking loops, and PID timing loops.

However, I have almost finished a dual PIC PID controller design, runnning
at 20MHz, and using interrupt driven bit-bashing to communicate between
PICs, using one PIC as the PID controller, and the other as the master.
This setup works really well.

PS: Olin - 1mF???  Isn't that just 1000uF?  Very common old garden variety
for an electtrolytic (I have a box full of them used for DC power
filtering), but I assume that for good LC filtering, an electro would not be
the best choice (too lossy).  Now, getting an MKT or mono in 1000uF size is
another matter ...

Rgs
Ian

> {Original Message removed}

2004\01\20@072624 by Olin Lathrop

face picon face
Ian McLean wrote:
> A combination of 100uH and 1mF seems good to me for an "off-the-shelf"
> solution, without making the inductor too big.  I cannot increase the
> PWM frequency much above 5kHz at this stage in the design I am using
> it in, because the PIC is only running at 4MHz.

That doesn't make sense.  A 4MHz clock results in a 1MHz instruction rate,
which must be divided by 64 to get 8 bit PWM resolution, which results in
15.6KHz PWM rate.  8 bit resolution is certainly enough for closed loop
control of a Peltier cooler.  Actually, 6 bits is probably more than
adequate.

> Unfortunately, increasing crystal speed will require a
> considerable amount of code revision to adjust timing, examples being
> PWM registers, serial registers, bit-bashing clocking loops, and PID
> timing loops.

This should not be the case for well written code.  You should always have
an assembly constant the specifies the clock speed, then everything else
computed from that.  Take a look at my UART_BAUD and TIMER2_USEC macros in
STD.INS.ASPIC at http://www.embedinc.com/pic for examples of how this can be
done.  Hard coding divider values and other timing parameters is just plain
bad practise.

> However, I have almost finished a dual PIC PID controller design,
> runnning at 20MHz, and using interrupt driven bit-bashing to
> communicate between PICs, using one PIC as the PID controller, and
> the other as the master. This setup works really well.
>
> PS: Olin - 1mF???  Isn't that just 1000uF?

The "m" stands for "milli" meaning 1/1000 and the "u" for "micro" meaning
1/1000000.  In engineering notation, one tries to keep the mantissa part so
that it has 1 to 3 digits left of the point.  While 1000uF does express the
value, 1mF is the preferred way to write it.

For some reason the standard engineering notation convention was not
followed for capacitance way back when.  The milli and nano ranges were
skipped so that everything was either micro or pico.  This led to cumbersome
values like .01uF or 10000pF.  Lately this sillyness is being abandoned.  I
make it a point to use proper engineering notation for values, including
capacitance.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservspamspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

2004\01\20@164223 by Ian McLean

flavicon
face
Hi Olin,

> That doesn't make sense.  A 4MHz clock results in a 1MHz
> instruction rate,
> which must be divided by 64 to get 8 bit PWM resolution,
> which results in
> 15.6KHz PWM rate.  8 bit resolution is certainly enough for
> closed loop
> control of a Peltier cooler.  Actually, 6 bits is probably more than
> adequate.
>

Oops, silly me.  How did I ever drop that leading '1' in my calculations,
I'll never know ;)


{Quote hidden}

Agreed, but this is a project "in the works", and did not start out as my
own code.   It is also large, over 6k of code space.  It does a lot more
than just PID control of Peltier's.  I will be going through the code and
pulling all of this stuff together once the basic implementation is
complete.  I am still using absolute mode, and not the linker, which does
not help ... but again, this is not originally my code so changing over from
absolute mode *will* require considerable revision.

> > However, I have almost finished a dual PIC PID controller design,
> > runnning at 20MHz, and using interrupt driven bit-bashing to
> > communicate between PICs, using one PIC as the PID controller, and
> > the other as the master. This setup works really well.

This one is completely my own design, done the way *I* like it.  It
implements all the points you have made out, and then some.

{Quote hidden}

Try telling this to parts suppliers, electronics ans schematical texts, and,
well, everywhere really.  I know my engineering notation very well thank you
Olin, but I am definately not used to seeing 1000uF written as 1mF, so I
thought it might have been a typo on your behalf ;)  I can understand the
"non-standardness" with capacitance however, because the standard measure
for 1 Farad is very big.  This problem is not restricted to capacitance
either.  Inductance and the Henry share the same problem.

Thanks and regards
Ian

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spamBeGonelistservKILLspamspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body

2004\01\20@172758 by Olin Lathrop

face picon face
Ian McLean wrote:
> This
> problem is not restricted to capacitance either.  Inductance and the
> Henry share the same problem.

But nobody seems to have an aversion to millihenrys.  I still don't
understand why milli and nano ever got skipped for capacitance values only,
and why some people still perpetuate that nonsense today.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservspam_OUTspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body

2004\01\21@072943 by Nigel Orr

flavicon
face
pic microcontroller discussion list wrote:
>>> PS: Olin - 1mF???  Isn't that just 1000uF?
>>
>> The "m" stands for "milli" meaning 1/1000 and the "u" for "micro"
>> meaning 1/1000000.  In engineering notation, one tries to keep the
>
> Try telling this to parts suppliers, electronics ans schematical
> texts, and, well, everywhere really.

I overheard one of my colleagues ordering some 47uF caps from a well known
distributor's call centre with little success, eventually saying "No,
MICRO, it looks like a small letter 'u'"...

I'll be charitable, assume it was a new employee at the distributors, and
not name the company :-)

IME, in the UK, pF,nF and uF are widely used, mF is rarely used.  It
doesn't help that 100uF is sometimes called '100 muff', which could equally
be 100 mF!

Nigel

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-request@spam@spammitvma.mit.edu

More... (looser matching)
- Last day of these posts
- In 2004 , 2005 only
- Today
- New search...