Homebrew PCBs

> <John.Craddock@...> wrote:
> >
> > Hi Philip,
> > Yes, I read the article and have been doing some research of
> thermocouple interfacing. ...

why so much complication?

hardware is dead simple using a 1-wire interface from Dallas-Maxim.
thermocouple feeds directly into an ADC (DS2761) that directly digitizes with a
resolution of 15uV, measures digitally its own temperature and connects to a uP
via one pin (and ground). also has a 2-5V ADC that can be fed directly from a
pot (ie, setpoint).

so, all is needed is this chip 
( http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3468 ) plus a microP (I use
Atmel Tiny), plus an opto-triac/triac pair or a solid state relay and a few
passives, switches and a pot with a large knob.

the process is all digital and the micro has the chance (responsibility) to
implement whatever algorithm fits (from on-off to self adjusting) plus any
bells and whistles (LCD display, serial logging, profile, etc).
BTW, it DOES fit on a 1" x 2" board (exclusive of power and relay).  

I kept following this discussion in the hope somebody would come up with a two
chip reasonable analog solution, but no, no such... (at least yet)





__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

Hi Herbert.

If I am reading it correctly, the DS2761 is a "High-Precision Li+ 
Battery Monitor". The datasheet says that its voltage measurement 
channel does so with a resolution of 4.88 millivolts resolution. Typical 
thermocouples have a resolution of around 40 to 50 microvolts per degree C.

So obviously I am missing something from your description below. Could 
you elaborate please?

Regards,

Ross McKenzie
ValuSoft
Melbourne Australia


Herbert E. Plett wrote:

>
>
> > <John.Craddock@...> wrote:
> > >
> > > Hi Philip,
> > > Yes, I read the article and have been doing some research of
> > thermocouple interfacing. ...
>
> why so much complication?
>
> hardware is dead simple using a 1-wire interface from Dallas-Maxim.
> thermocouple feeds directly into an ADC (DS2761) that directly 
> digitizes with a
> resolution of 15uV, measures digitally its own temperature and 
> connects to a uP
> via one pin (and ground). also has a 2-5V ADC that can be fed directly 
> from a
> pot (ie, setpoint).
>
> so, all is needed is this chip
> ( http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3468 
> <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3468> ) plus a microP 
> (I use
> Atmel Tiny), plus an opto-triac/triac pair or a solid state relay and 
> a few
> passives, switches and a pot with a large knob.
>
> the process is all digital and the micro has the chance 
> (responsibility) to
> implement whatever algorithm fits (from on-off to self adjusting) plus any
> bells and whistles (LCD display, serial logging, profile, etc).
> BTW, it DOES fit on a 1" x 2" board (exclusive of power and relay).
>
> I kept following this discussion in the hope somebody would come up 
> with a two
> chip reasonable analog solution, but no, no such... (at least yet)
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com <http://mail.yahoo.com>
>
>

On Thu, 24 Aug 2006 14:48:12 +0200, Herbert E. Plett <cachureos@...>  
wrote:

>
> why so much complication?
> hardware is dead simple using a 1-wire interface from Dallas-Maxim.
> thermocouple feeds directly into an ADC (DS2761) that directly digitizes  
> with a
> resolution of 15uV, measures digitally its own temperature and connects  
> to a uP
> via one pin (and ground). also has a 2-5V ADC that can be fed directly  
> from a
> pot (ie, setpoint).
> so, all is needed is this chip
> ( http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3468 ) plus a microP (I  
> use
> Atmel Tiny), plus an opto-triac/triac pair or a solid state relay and a  
> few
> passives, switches and a pot with a large knob.
> the process is all digital and the micro has the chance (responsibility)  
> to
> implement whatever algorithm fits (from on-off to self adjusting) plus  
> any
> bells and whistles (LCD display, serial logging, profile, etc).
> BTW, it DOES fit on a 1" x 2" board (exclusive of power and relay).
> I kept following this discussion in the hope somebody would come up with  
> a two
> chip reasonable analog solution, but no, no such... (at least yet)


I bet you can't do it under 20eur either...

ST

--- In Homebrew_PCBs@yahoogroups.com, "Herbert E. Plett"
<cachureos@...> wrote:
>
> 
> 
> > <John.Craddock@> wrote:
> > >
> > > Hi Philip,
> > > Yes, I read the article and have been doing some research of
> > thermocouple interfacing. ...
> 
> why so much complication?
> 
> hardware is dead simple using a 1-wire interface from Dallas-Maxim.
> thermocouple feeds directly into an ADC (DS2761) that directly
digitizes with a
> resolution of 15uV, measures digitally its own temperature and
connects to a uP
> via one pin (and ground). also has a 2-5V ADC that can be fed
directly from a
> pot (ie, setpoint).
> 
> so, all is needed is this chip 
> ( http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3468 ) plus a microP
(I use
> Atmel Tiny), plus an opto-triac/triac pair or a solid state relay
and a few
> passives, switches and a pot with a large knob.
> 
> the process is all digital and the micro has the chance
(responsibility) to
> implement whatever algorithm fits (from on-off to self adjusting)
plus any
> bells and whistles (LCD display, serial logging, profile, etc).
> BTW, it DOES fit on a 1" x 2" board (exclusive of power and relay).  
> 
> I kept following this discussion in the hope somebody would come up
with a two
> chip reasonable analog solution, but no, no such... (at least yet)
> 
> 
> 
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com
>


Hi Herbert.

If I am reading it correctly, the DS2761 is a "High-Precision Li+
Battery Monitor". The datasheet says that its voltage measurement
channel does so with a resolution of 4.88 millivolts resolution.
Typical thermocouples have a resolution of around 40 to 50 microvolts
per degree C.

So obviously I am missing something from your description below. Could
you elaborate please?

Regards,

Ross McKenzie
ValuSoft
Melbourne Australia

Maybe MAX6675? Thermocouple input with built-in cold junction
compensation, 12bit ADC.

BTW, this needs to move to Electronics_101. I've posted a rough
schematic/block diagram of a potential circuit using a Silicon diode
for cold junction compensation, with a simple Wheatstone bridge
circuit to use your DMM as a thermometer, only active part is a
voltage regulator. 1mV/C.

http://groups.yahoo.com/group/Electronics_101/

Steve Greenfield

Or maybe: -
http://groups.yahoo.com/group/E-Z_Bake/

This thread is related to Stefan's toaster oven mods to be able to bake direct inkjet pcbs; and perhaps to double as a reflow oven (my objective). BTW, I bought the MAX6675 (A$8.64), much happier now, thanks to help from members of this list. BTW2 a diode compensator requires a isothermal block at the cold junction; the MAX6675 gets away without this.
Regards
John C

> -----Original Message-----
> From: Steve [mailto:alienrelics@...]
> Sent: 25 August 2006 07:13
> To: Homebrew_PCBs@yahoogroups.com
> Subject: [Homebrew_PCBs] Re: How to DO thermocouple interfacing
> BTW, this needs to move to Electronics_101. I've posted a rough
> schematic/block diagram of a potential circuit using a Silicon diode
> for cold junction compensation, with a simple Wheatstone bridge
> circuit to use your DMM as a thermometer, only active part is a
> voltage regulator. 1mV/C.
> 
> http://groups.yahoo.com/group/Electronics_101/
> 
> Steve Greenfield
> 
> 
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, 
> Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links
> 
> 
> 
>  
> 
> 
> 
>

> ST wrote:
> I bet you can't do it under 20eur either...

I bet I can do it for under $30 AUD* :P (with
the LCD screen)


*The $30 AUD would be the prices I get
components for not the dicky smith / jaycar /
radio spares price of parts

--- In Homebrew_PCBs@yahoogroups.com, "John Craddock"
<John.Craddock@...> wrote:
>
> Or maybe: -
> http://groups.yahoo.com/group/E-Z_Bake/
> 
> This thread is related to Stefan's toaster oven mods to be able to
> bake direct inkjet pcbs; and perhaps to double as a reflow oven
> (my objective). 

Yes, but it is squarely into electronic design now. We've been over
this before, if you try and say that anything remotely related to
making PCBs is on topic, soon we're discussing brake linings and
baseball scores because you might drive your car to pick up hydrogen
peroxide and listen to the game on the way.

There have already been past discussions and links to PID and such on
the Electronics_101 list. Absolutely integral to a project such as this.

> BTW, I bought the MAX6675 (A$8.64), much happier now,
> thanks to help from members of this list. BTW2 a diode
> compensator requires a isothermal block at the cold junction;
> the MAX6675 gets away without this.

I just ordered a couple of samples of both the MAX6675 and DS2761.
Although the DS2761 is a LiIon battery monitor IC, Maxim has an app
note on how to use it as a thermocouple chip. The thermocouple
connects to the battery voltage monitor input, and the chip's built-in
temp monitors cold junction temp. However, an external computer must
do all the processing.

No, the MAX6675 does -not- remove the need for some kind of isothermal
block. The thermocouple wires -must- be soldered/connected right at
the IC, or the connections will be at a different temp than the sensor
inside the IC.

Having a silicon diode in intimate physical contact with the
thermocouple to PCB connections is not difficult.

Steve Greenfield

Steve, Sorry to tread on your toes! If you read my post carefully you will note an alternative suggestion that concurs with your original point. I consider that the following parts of the post supports the context of the suggestion of the EZ-bake group.
Regards
John C

> -----Original Message-----
> From: Steve [mailto:alienrelics@...]
> Sent: 25 August 2006 08:29
> To: Homebrew_PCBs@yahoogroups.com
> Subject: [Homebrew_PCBs] Re: How to DO thermocouple interfacing
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, "John Craddock"
> <John.Craddock@...> wrote:
> >
> > Or maybe: -
> > http://groups.yahoo.com/group/E-Z_Bake/
> > 
> > This thread is related to Stefan's toaster oven mods to be able to
> > bake direct inkjet pcbs; and perhaps to double as a reflow oven
> > (my objective). 
> 
> Yes, but it is squarely into electronic design now. We've been over
> this before, if you try and say that anything remotely related to
> making PCBs is on topic, soon we're discussing brake linings and
> baseball scores because you might drive your car to pick up hydrogen
> peroxide and listen to the game on the way.
> 
> There have already been past discussions and links to PID and such on
> the Electronics_101 list. Absolutely integral to a project 
> such as this.
> 
> > BTW, I bought the MAX6675 (A$8.64), much happier now,
> > thanks to help from members of this list. BTW2 a diode
> > compensator requires a isothermal block at the cold junction;
> > the MAX6675 gets away without this.
> 
> I just ordered a couple of samples of both the MAX6675 and DS2761.
> Although the DS2761 is a LiIon battery monitor IC, Maxim has an app
> note on how to use it as a thermocouple chip. The thermocouple
> connects to the battery voltage monitor input, and the chip's built-in
> temp monitors cold junction temp. However, an external computer must
> do all the processing.
> 
> No, the MAX6675 does -not- remove the need for some kind of isothermal
> block. The thermocouple wires -must- be soldered/connected right at
> the IC, or the connections will be at a different temp than the sensor
> inside the IC.
> 
> Having a silicon diode in intimate physical contact with the
> thermocouple to PCB connections is not difficult.
> 
> Steve Greenfield
> 
> 
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, 
> Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links
> 
> 
> 
>  
> 
> 
>

don't get fooled by the title.
the DS2761 chip has two voltage inputs, one resolves 15uV in +-32mV (direct
connection of thermocouple up to 500C) and the other resolves 4.8mV in 2-5V
range (good for a pot input). on top of this it measures it's own temperature
and has an i/o pin (plus a switch sensor).
the setup is very simple: the thermocouple goes to a screw terminal close to
the chip (isothermal) and the chip goes to one pin of the uC. the uC doesn't
even need a crystal, so the most expensive parts are the ref pot, the relay,
the power supply and other stuff you may add (LCD...), BTW all of this also
needed with an analog design.
the algorithm is also basically simple:
read the internal temp
read the uV input and scale to temp units
read the V input and scale to temp units
compare V to uV+T and make the on-off decision (the actual control algorithm)
set the relay accordingly
repeat once a second...

works like a charm, but I do only on/off (thermostat). an 'intelligent' control
algorithm is out of my scope as I have no (mathematical) control theory
background.
some smart procedure can be implemented to input the profile using switches for
command sequence and the pot for values if you want a stand-alone control w/o a
PC link.



--- Ross McKenzie <valusoft@...> wrote:
> If I am reading it correctly, the DS2761 is a "High-Precision Li+ 
> Battery Monitor". The datasheet says that its voltage measurement 
> channel does so with a resolution of 4.88 millivolts resolution. Typical 
> thermocouples have a resolution of around 40 to 50 microvolts per degree C.
> 
> So obviously I am missing something from your description below. Could 
> you elaborate please?


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

Thanks.

So you are suggesting using the "IS1" and "IS2" pins without any current 
sense resistor (internal or external). Is that correct?

Regards,

Ross

Herbert E. Plett wrote:

> don't get fooled by the title.
> the DS2761 chip has two voltage inputs, one resolves 15uV in +-32mV 
> (direct
> connection of thermocouple up to 500C) and the other resolves 4.8mV in 
> 2-5V
> range (good for a pot input). on top of this it measures it's own 
> temperature
> and has an i/o pin (plus a switch sensor).
> the setup is very simple: the thermocouple goes to a screw terminal 
> close to
> the chip (isothermal) and the chip goes to one pin of the uC. the uC 
> doesn't
> even need a crystal, so the most expensive parts are the ref pot, the 
> relay,
> the power supply and other stuff you may add (LCD...), BTW all of this 
> also
> needed with an analog design.
> the algorithm is also basically simple:
> read the internal temp
> read the uV input and scale to temp units
> read the V input and scale to temp units
> compare V to uV+T and make the on-off decision (the actual control 
> algorithm)
> set the relay accordingly
> repeat once a second...
>
> works like a charm, but I do only on/off (thermostat). an 
> 'intelligent' control
> algorithm is out of my scope as I have no (mathematical) control theory
> background.
> some smart procedure can be implemented to input the profile using 
> switches for
> command sequence and the pot for values if you want a stand-alone 
> control w/o a
> PC link.
>
> --- Ross McKenzie <valusoft@... 
> <mailto:valusoft%40optusnet.com.au>> wrote:
> > If I am reading it correctly, the DS2761 is a "High-Precision Li+
> > Battery Monitor". The datasheet says that its voltage measurement
> > channel does so with a resolution of 4.88 millivolts resolution. 
> Typical
> > thermocouples have a resolution of around 40 to 50 microvolts per 
> degree C.
> >
> > So obviously I am missing something from your description below. Could
> > you elaborate please?
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com <http://mail.yahoo.com>
>
>

yep, just a cap to keep out noise and two high valued resistors to force open
TC into infinity.
I just chose the polarity so as to make the scaling easier. I use J type TC and
'linearize' it into two ranges instead of applying the 8 term polynomial (too
many instructions).



--- Ross McKenzie <valusoft@...> wrote:
> So you are suggesting using the "IS1" and "IS2" pins without any current 
> sense resistor (internal or external). Is that correct?
>

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

...and the two high valued resistors, eg 10Mohm, one from ground to IS2 
and the other from IS2 to supply rail?


Herbert E. Plett wrote:

> yep, just a cap to keep out noise and two high valued resistors to 
> force open
> TC into infinity.
> I just chose the polarity so as to make the scaling easier. I use J 
> type TC and
> 'linearize' it into two ranges instead of applying the 8 term 
> polynomial (too
> many instructions).
>
> --- Ross McKenzie <valusoft@... 
> <mailto:valusoft%40optusnet.com.au>> wrote:
> > So you are suggesting using the "IS1" and "IS2" pins without any 
> current
> > sense resistor (internal or external). Is that correct?
> >
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com <http://mail.yahoo.com>
>
>

For a two chip analog solution, why not the Linear Technology LT1025 -
cold compensation, and separate inputs for type J, K or T and R or S.
Digikey has it for $6 Can. To amplify add a LTC1052 chopper amp ($10 C). 

Go to:

http://www.dtweed.com/circuitcellar/caj00031.htm

and look at the PDF for "Temperature Monitoring" by Steve Ciarcia.

Grant

--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan"
<stefan_trethan@...> wrote:
>
> On Thu, 24 Aug 2006 14:48:12 +0200, Herbert E. Plett <cachureos@...>  
> wrote:
> 
> >
> > why so much complication?
> > hardware is dead simple using a 1-wire interface from Dallas-Maxim.
> > thermocouple feeds directly into an ADC (DS2761) that directly
digitizes  
> > with a
> > resolution of 15uV, measures digitally its own temperature and
connects  
> > to a uP
> > via one pin (and ground). also has a 2-5V ADC that can be fed
directly  
> > from a
> > pot (ie, setpoint).
> > so, all is needed is this chip
> > ( http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3468 ) plus a
microP (I  
> > use
> > Atmel Tiny), plus an opto-triac/triac pair or a solid state relay
and a  
> > few
> > passives, switches and a pot with a large knob.
> > the process is all digital and the micro has the chance
(responsibility)  
> > to
> > implement whatever algorithm fits (from on-off to self adjusting)
plus  
> > any
> > bells and whistles (LCD display, serial logging, profile, etc).
> > BTW, it DOES fit on a 1" x 2" board (exclusive of power and relay).
> > I kept following this discussion in the hope somebody would come
up with  

> > a two
> > chip reasonable analog solution, but no, no such... (at least yet)
> 
> 
> I bet you can't do it under 20eur either...
> 
> ST
>