EDAboard.com | EDAboard.eu | EDAboard.de | EDAboard.co.uk | RTV forum PL | NewsGroups PL

Temperature compensating diode drops

Ask a question - edaboard.com

elektroda.net NewsGroups Forum Index - Electronics Design - Temperature compensating diode drops

Goto page Previous  1, 2, 3, 4  Next

Jim Thompson
Guest

Sun Jan 08, 2017 4:49 am   



On Fri, 06 Jan 2017 16:32:54 -0700, Jim Thompson
<To-Email-Use-The-Envelope-Icon_at_On-My-Web-Site.com> wrote:

Quote:
On Fri, 6 Jan 2017 03:42:06 -0500, Phil Hobbs
pcdhSpamMeSenseless_at_electrooptical.net> wrote:
[snip]

I've sometimes got round that by putting a Schottky diode in series with
the sense resistor, like this. (The base current for Q1 actually comes
from a RRO op amp via R1--it's a class B complementary bridge.)

Q1
0-*--------- -----*---->|---R2R2---*------0
| \ A | |
| ------ ------- |
| | / V Q2 |
*--R1R1-------*---- -------------*

[snip]

I'm not totally following your scheme, but wouldn't you want to
_replace_Q2_ with the TLV431 (using drawing _above_)? Then your
current limit is 1.2V/R2 with a _very_low_ TC. A TLV431 is
transistor-sized... TO-92 pack, unless you're in love with SOT's.


[snip]


Depending on the voltages involved... and the switching speed... you
might simply make the bridge switches part of the current control.

More info about what you're attempting?

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

piglet
Guest

Sun Jan 08, 2017 5:46 am   



On 07/01/2017 22:20, pcdhobbs_at_gmail.com wrote:
Quote:
Depending on the voltages involved... and the switching speed... you
might simply make the bridge switches part of the current control.

More info about what you're attempting?

It's a fairly simple bridge driver for a thermoelectric cooler. Normally I'd put a single current limiter in series with the input, but in this case I need different current limits for positive and negative.

The most recent iteration has a class B current conveyor on the left side, driven by an op amp. The right side is voltage-driven, so that the outputs are centred on Vcc/2.

Because of the asymmetric limits, I need at least two separate current limiters, one for each polarity. Seems like I only need to limit one side, right?

However, if the left side of the TEC ever got shorted to ground, the right side could supply enough current to turn it to lava in short order. The lasers are expensive, not to mention service calls and down time, so it seems like a good idea to put current limits in all four legs of the bridge.

(Class D is unattractive due to noise coupling into the laser.)

Cheers

Phil Hobbs


Now I see why you want 4 instead of 2. But I think only 2 limiters need
be semi-precise as per topic and the other 2 limiters can be higher
current and coarser without Vbe compensation as they only need to limit
s/c fault current?

piglet

Jim Thompson
Guest

Sun Jan 08, 2017 5:59 am   



On Sat, 7 Jan 2017 14:20:57 -0800 (PST), pcdhobbs_at_gmail.com wrote:

Quote:
Depending on the voltages involved... and the switching speed... you
might simply make the bridge switches part of the current control.

More info about what you're attempting?

It's a fairly simple bridge driver for a thermoelectric cooler. Normally I'd put a single current limiter in series with the input, but in this case I need different current limits for positive and negative.

The most recent iteration has a class B current conveyor on the left side, driven by an op amp. The right side is voltage-driven, so that the outputs are centred on Vcc/2.

Because of the asymmetric limits, I need at least two separate current limiters, one for each polarity. Seems like I only need to limit one side, right?


Won't asymmetric current limits screw with the symmetry implied by
voltage drive?

Quote:

However, if the left side of the TEC ever got shorted to ground, the right side could supply enough current to turn it to lava in short order. The lasers are expensive, not to mention service calls and down time, so it seems like a good idea to put current limits in all four legs of the bridge.

(Class D is unattractive due to noise coupling into the laser.)

Cheers

Phil Hobbs


Seems to me a gross limiter up top for shorts to ground, then limits
on each lower leg for current control.

If you want to share your schematic off-line I'll provide free advice
;-)

(I'm having trouble envisioning/imagining the current conveyor
configuration Smile

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

Jim Thompson
Guest

Sun Jan 08, 2017 10:57 pm   



On Fri, 6 Jan 2017 16:23:04 -0800 (PST), pcdhobbs_at_gmail.com wrote:

Quote:
I'm not totally following your scheme, but wouldn't you want to
_replace_Q2_ with the TLV431 (using drawing _above_)? áThen your
current limit is 1.2V/R2 with a _very_low_ TC. áA TLV431 is
transistor-sized... TO-92 pack, unless you're in love with SOT's.

Yup. The LM385-ADJ is the same sort of thing except that it holds the 1.2V between ADJ and K, so it works with PNPs. (To make a 1.2V reference with a TLV431 you short ADJ to K, whereas with an LM385 it's ADJ to A.)

Still costs an extra volt-and-a-bit of headroom, and another buck or so on the BOM. Might be worth it.

Cheers

Phil Hobbs


I didn't have in mind tying ADJ to K, etc, I had in mind some
variation of this...

<http://www.analog-innovations.com/SED/PCDH_CL_2017-01-08.png>

I can envision various switching schemes where this structure becomes
part of the H-bridge.

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |


Guest

Mon Jan 09, 2017 12:34 am   



Quote:
That runs into money, though.  Negative LDOs especially.


Quote:
Why negative LDOs? You'll have to look for the LM317-kind of
three-legged regulators. Resistor in series, ADJ terminal to other side,
done. Since you need it to work in both directions you'll have to hang a
Schottky across and then another such (positive) regulator in series,
also with a Schottky. Except that one is reversed. Not much money.


Thanks.

I misunderstood--you're talking about putting them in series with the TEC rather than using them directly as the pass elements in the bridge.

Problem is, there's a cable in between, so if there's a short to ground it has a 50:50 chance of being on the wrong side of the current limiter if there's only one.

Thus I'd need one set on each side, which is more complicated than what I'm doing already.

Cheers

Phil Hobbs


Guest

Mon Jan 09, 2017 12:54 am   



Quote:
None here, the world is using too much antibiotics as it is. On a long
flight to a client I must have contracted a bug and it hit me hard two
days later. Their VC of Quality Control recommended a "guaranteed fix":
A bottle of Southern Comfort. It worked. I didn't drink the whole bottle
though.


Yikes, Southern Comfort. Vanilla extract is about the same price and probably a lot more palatable. ;)

Of course the prize here goes to Mary Hartman, Mary Hartman (1976). Her next door neighbour, Coach Fedder, was treating a cold with large quantities of bourbon. She brought him a large bowl of chicken soup, and he promptly passed out and drowned in it. (I'm not a huge Norman Lear fan in general, but he was undeniably a genius.)

Cheers

Phil Hobbs

Joerg
Guest

Mon Jan 09, 2017 1:11 am   



On 2017-01-06 00:42, Phil Hobbs wrote:
Quote:
Hi, all,

I'm designing a diode laser controller board for a customer in
Scandinavia. It has a sub-Poissonian adjustable current supply, a
resistor-linearized thermistor temperature sensor, and of course a
thermoelectric cooler driver.

The TEC requires different current limits in different quadrants. (A
TEC is actually a four-quadrant device since it generates electrical
power from the temperature difference as well as the other way round.)

With a bridged current driver, that can run into quite a few parts--you
need one current limiter per polarity, or if you want any short circuit
protection, one per leg (4 in all). So simplicity is at a bit of a
premium.

The usual two-terminal, two-BJT current limiter suffers from a fairly
gross temperature coefficient, like 3000 ppm/K. Protecting a TEC isn't
a super high precision application, but something a bit better than that
would be good.

I've sometimes got round that by putting a Schottky diode in series with
the sense resistor, like this. (The base current for Q1 actually comes
from a RRO op amp via R1--it's a class B complementary bridge.)

Q1
0-*--------- -----*---->|---R2R2---*------0
| \ A | |
| ------ ------- |
| | / V Q2 |
*--R1R1-------*---- -------------*

Trouble is, Schottkies don't have the same dV/dT as BJTs--it's more like
1 mV/K at high current, vs. 2 mV/K at low current, so you don't gain
that much--about 1500 ppm/K.

A PN diode connected as above plus something like a TLV431 in series
with Q2's emitter will do a good job, probably 300-500 ppm/V, which
would be fine. However, that costs four more ICs and another 2.4V of
headroom (1.2V per side in each polarity).

Seems like an arbitrage opportunity. ;)

My current thinking is to put the diode in series with the base of Q2
and use Q1's V_BE to provide a reasonably stable bias current, like this:

Q1 1ohm
0-*--------- --------*-------R2R2-----*------0
| \ A | |
| ------ A Schottky |
| | | |
*---R1R1------*--R3R3----* |
| | |
| ------ |
| / V Q2 |
*------- --------------*

The diode runs at much lower current, and so has a higher dV/dT, and by
choosing R2 correctly, I can adjust it to match Q2's a bit better.

One wouldn't want to bet one's firstborn child on the unit-to-unit
consistency being too wonderful, but ISTM that I ought to be able to get
a factor of maybe 6 this way.

Anybody else done this sort of thing? Better suggestions welcome!


Why not use linear regulators as current sources? If the total drop is
too high you could look for an LDO that can be pressed into such
service. This can reduce the BOM to two parts per leg.

Alternatively build the limiter into your driver and use a comparators
as quadrant detectors to switch gears on the limiter.

--
Regards, Joerg

http://www.analogconsultants.com/

Phil Hobbs
Guest

Mon Jan 09, 2017 1:39 am   



On 01/08/2017 01:11 PM, Joerg wrote:
Quote:
On 2017-01-06 00:42, Phil Hobbs wrote:
Hi, all,

I'm designing a diode laser controller board for a customer in
Scandinavia. It has a sub-Poissonian adjustable current supply, a
resistor-linearized thermistor temperature sensor, and of course a
thermoelectric cooler driver.

The TEC requires different current limits in different quadrants.
(A TEC is actually a four-quadrant device since it generates
electrical power from the temperature difference as well as the
other way round.)

With a bridged current driver, that can run into quite a few
parts--you need one current limiter per polarity, or if you want
any short circuit protection, one per leg (4 in all). So
simplicity is at a bit of a premium.

The usual two-terminal, two-BJT current limiter suffers from a
fairly gross temperature coefficient, like 3000 ppm/K. Protecting
a TEC isn't a super high precision application, but something a bit
better than that would be good.

I've sometimes got round that by putting a Schottky diode in series
with the sense resistor, like this. (The base current for Q1
actually comes from a RRO op amp via R1--it's a class B
complementary bridge.)

Q1 0-*--------- -----*---->|---R2R2---*------0 | \
A | | | ------ ------- | |
| / V Q2 | *--R1R1-------*----
-------------*

Trouble is, Schottkies don't have the same dV/dT as BJTs--it's more
like 1 mV/K at high current, vs. 2 mV/K at low current, so you
don't gain that much--about 1500 ppm/K.

A PN diode connected as above plus something like a TLV431 in
series with Q2's emitter will do a good job, probably 300-500
ppm/V, which would be fine. However, that costs four more ICs and
another 2.4V of headroom (1.2V per side in each polarity).

Seems like an arbitrage opportunity. ;)

My current thinking is to put the diode in series with the base of
Q2 and use Q1's V_BE to provide a reasonably stable bias current,
like this:

Q1 1ohm 0-*---------
--------*-------R2R2-----*------0 | \ A |
| | ------ A Schottky | | |
| | *---R1R1------*--R3R3----* | |
| | | ------ | | / V Q2
| *------- --------------*

The diode runs at much lower current, and so has a higher dV/dT,
and by choosing R2 correctly, I can adjust it to match Q2's a bit
better.

One wouldn't want to bet one's firstborn child on the unit-to-unit
consistency being too wonderful, but ISTM that I ought to be able
to get a factor of maybe 6 this way.

Anybody else done this sort of thing? Better suggestions welcome!


Why not use linear regulators as current sources? If the total drop
is too high you could look for an LDO that can be pressed into such
service. This can reduce the BOM to two parts per leg.


That runs into money, though. Negative LDOs especially. I have samples
of the laser/TEC combinations, so once I get back on my feet I'll do
some measurements and see if I can afford an extra volt-and-a-bit for
the TLV431/LM4041-ADJ version.

Quote:

Alternatively build the limiter into your driver and use a
comparators as quadrant detectors to switch gears on the limiter.


That might be interesting. The issue is that neither side of the TEC is
grounded, and it's on the other end of a cable, which makes it somewhat
vulnerable to having one side shorted to ground. That makes any
asymmetrical on-board solution a bit problematic--the TEC can melt
before the thermistor notices anything wrong.

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

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

Phil Hobbs
Guest

Mon Jan 09, 2017 1:55 am   



On 01/08/2017 10:57 AM, Jim Thompson wrote:
Quote:
On Fri, 6 Jan 2017 16:23:04 -0800 (PST), pcdhobbs_at_gmail.com wrote:

I'm not totally following your scheme, but wouldn't you want to
_replace_Q2_ with the TLV431 (using drawing _above_)? Then your
current limit is 1.2V/R2 with a _very_low_ TC. A TLV431 is
transistor-sized... TO-92 pack, unless you're in love with SOT's.

Yup. The LM385-ADJ is the same sort of thing except that it holds

the 1.2V between ADJ and K, so it works with PNPs. (To make a 1.2V
reference with a TLV431 you short ADJ to K, whereas with an LM385 it's
ADJ to A.)
Quote:

Still costs an extra volt-and-a-bit of headroom, and another buck or
so on the BOM. Might be worth it.


My previous followup seems to have gone off into the Howling Ether.

Quote:
I didn't have in mind tying ADJ to K, etc, I had in mind some
variation of this...

http://www.analog-innovations.com/SED/PCDH_CL_2017-01-08.png


Yes, exactly--it's an IC-ified version of the normal 2-terminal limiter.
The 431-style refs only work with NPNs and NFETs, though, because they
hold Vref between ADJ and A. For PNPs you need something upside-down
such as the LM385 or LM4041-ADJ, which hold Vref between K and ADJ.

Quote:
I can envision various switching schemes where this structure becomes
part of the H-bridge.


Yup, that's the idea, except that mine is linear. The maximum current
is +-250 mA, so the dissipation is modest, and this avoids problems
with switching crap getting into the laser.

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

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


Guest

Mon Jan 09, 2017 2:15 am   



Quote:
Problem is, there's a cable in between, so if there's a short to
ground it has a 50:50 chance of being on the wrong side of the
current limiter if there's only one.


I thought you only needed to protect the TEC, not the driver.


Right. But the driver is a bridge, and a short to ground on the wrong side will do b ad things if the protection is asymmetric.

Quote:
Thus I'd need one set on each side, which is more complicated than
what I'm doing already.


Can't the driver be beefed up a bit and adequately fused by electronics
means so it won't smoke?


'Tain't the driver I'm worried about, it's the TEC. Symmetric drive over a cable requires symmetric protection. If the current limits were the same for positive and negative drive, it would be easy, but they aren't.

Cheers

Phil Hobbs


Guest

Mon Jan 09, 2017 2:36 am   



Quote:
Should be easy enough to convey to the "conveyor" what
quadrant it's in... n'est-ce pas ?:


A current conveyor is a current in/current out gain stage. This one is class B, so it's pretty simple.

The driver stage is common base, and the output is common emitter, connected to the supply rail with emitter degeneration.

Thus neglecting Vbe drops (which is okay since the application is low speed and there are op amps wrapped round), a milliamp in produces N milliamps out. You put NPN and PNP versions with their inputs and outputs in parallel, and you have a current conveyor--low Z in, high Z out.

It's the fourth member of the family: normal op amps (voltage in, voltage out), CFB amps (current in, voltage out), OTAs (voltage in, current out), and current conveyors (current in, current out).

Cheers

Phil Hobbs


Guest

Mon Jan 09, 2017 2:46 am   



Quote:
Maybe I am understanding it wrong but if the current limiters are
directly in line with and physically at the Peltier it shouldn't matter.


Right, except that they aren't. The current limiters are on the board, and the laser/TEC is at the other end of a cable. Otherwise I wouldn't be worried about shorts.

>You could have one current limit for cool and another for heat. Of

That's the point of the exercise, right.

Quote:
course, if there is a chance that a wrench falls directly onto the
Peltier contacts that's bad.


If the laser assembly could be mounted on the board, I wouldn't be worried, but it can't. The application is an industrial tester for manufacturing other things, so reliability is key because down time is expensive.

Not all of the users are in the first world, so "expensive" has to take account of the BOM cost as well. If it were like Ford and GM, I'd throw a couple of bucks' BOM cost at it and move on.

Cheers

Phil Hobbs

Jim Thompson
Guest

Mon Jan 09, 2017 2:59 am   



On Sun, 8 Jan 2017 13:55:32 -0500, Phil Hobbs
<pcdhSpamMeSenseless_at_electrooptical.net> wrote:

Quote:
On 01/08/2017 10:57 AM, Jim Thompson wrote:
On Fri, 6 Jan 2017 16:23:04 -0800 (PST), pcdhobbs_at_gmail.com wrote:

I'm not totally following your scheme, but wouldn't you want to
_replace_Q2_ with the TLV431 (using drawing _above_)? Then your
current limit is 1.2V/R2 with a _very_low_ TC. A TLV431 is
transistor-sized... TO-92 pack, unless you're in love with SOT's.

Yup. The LM385-ADJ is the same sort of thing except that it holds
the 1.2V between ADJ and K, so it works with PNPs. (To make a 1.2V
reference with a TLV431 you short ADJ to K, whereas with an LM385 it's
ADJ to A.)

Still costs an extra volt-and-a-bit of headroom, and another buck or
so on the BOM. Might be worth it.

My previous followup seems to have gone off into the Howling Ether.


I didn't have your @gmail address white-listed Sad It is now.

(I have various @gmail/@google posters whitelisted followed by a
blanket kill of all the rest :-)

Quote:

I didn't have in mind tying ADJ to K, etc, I had in mind some
variation of this...

http://www.analog-innovations.com/SED/PCDH_CL_2017-01-08.png

Yes, exactly--it's an IC-ified version of the normal 2-terminal limiter.
The 431-style refs only work with NPNs and NFETs, though, because they
hold Vref between ADJ and A. For PNPs you need something upside-down
such as the LM385 or LM4041-ADJ, which hold Vref between K and ADJ.

I can envision various switching schemes where this structure becomes
part of the H-bridge.

Yup, that's the idea, except that mine is linear. The maximum current
is +-250 mA, so the dissipation is modest, and this avoids problems
with switching crap getting into the laser.

Cheers

Phil Hobbs


If it's linear, wouldn't it be easier to merge limits into the
"conveyor"?

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

Jim Thompson
Guest

Mon Jan 09, 2017 3:02 am   



On Sun, 8 Jan 2017 13:39:00 -0500, Phil Hobbs
<pcdhSpamMeSenseless_at_electrooptical.net> wrote:


[snip]

Quote:
so once I get back on my feet I'll do
some measurements and see if I can afford an extra volt-and-a-bit for
the TLV431/LM4041-ADJ version.

[snip]


Have you suffered an injury, or a surgery?

I'm considering having the joint in my right big toe replaced. I'm a
"walker" and the arthritis is impending my speed Wink

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

Jim Thompson
Guest

Mon Jan 09, 2017 3:06 am   



On Sun, 08 Jan 2017 13:02:27 -0700, Jim Thompson
<To-Email-Use-The-Envelope-Icon_at_On-My-Web-Site.com> wrote:

Quote:
On Sun, 8 Jan 2017 13:39:00 -0500, Phil Hobbs
pcdhSpamMeSenseless_at_electrooptical.net> wrote:


[snip]

so once I get back on my feet I'll do
some measurements and see if I can afford an extra volt-and-a-bit for
the TLV431/LM4041-ADJ version.

[snip]

Have you suffered an injury, or a surgery?

I'm considering having the joint in my right big toe replaced. I'm a
"walker" and the arthritis is impending my speed Wink

...Jim Thompson


Make that impeding... spell-checkers Sad

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

Goto page Previous  1, 2, 3, 4  Next

elektroda.net NewsGroups Forum Index - Electronics Design - Temperature compensating diode drops

Ask a question - edaboard.com

Arabic versionBulgarian versionCatalan versionCzech versionDanish versionGerman versionGreek versionEnglish versionSpanish versionFinnish versionFrench versionHindi versionCroatian versionIndonesian versionItalian versionHebrew versionJapanese versionKorean versionLithuanian versionLatvian versionDutch versionNorwegian versionPolish versionPortuguese versionRomanian versionRussian versionSlovak versionSlovenian versionSerbian versionSwedish versionTagalog versionUkrainian versionVietnamese versionChinese version
RTV map EDAboard.com map News map EDAboard.eu map EDAboard.de map EDAboard.co.uk map