level shifter

Yep. Always a test of my endurance to bow lower than they did ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

You should be able to make something along this line work satisfactorily. I= t is more compact than a diff amp or current source with inductor, but ther= e's no cost savings with DK's pricing. It is standard current source level = shift except it is voltage controlled by the 3.3V so the shift tracks with = power supply drift. Please view in a fixed-width font such as Courier.

. . . - . Q >----+-----. 3.3V GND . | | | | . | | | '->| . | R [R] |--. . C --- [ - ] | /| | . --- 3 '>| | | . | | |-----+----' . | | /| | . | | | | . '-----+-------------------------> GATE . | | | . | | | . | .---+ | . | | | | . \| | |/ 5 . |--+-| [- x R] . 3 . | | | . | | | . | | | . ------+---+----+-------+--- . | . -5V . . .

.

As shown in...

...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

It takes only elementary math to do it exactly, as in...

Monkeying with it, and the loop compensation, I suspect it would be simpler to drive the current mirrors from the emitters (common base). ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

That is inarguably more precise but it uses a pile of resistors, and all hi= s power supplies are for logic so no precision there for a reference. I don= 't think the pHEMT drive is all that critical. I realize now that he said = the output mean is about 1V down from the 3.3 so he wants to shift it from =

2.3 to -.4 for 2.7 total, means the R/3 should be 2.7/3.3 x R and not R/3.

more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

The Fairchild thing is four separate transistors in one epoxy package. Thermally, it's not going to be much better than separate transistors. You're counting on two current mirrors to be accurate. I don't trust current mirrors for accuracy, especially when made from discrete parts.

And the current through the level-shift resistor depends on the difference in the voltages of both power supplies. Probably the tie on the lower right should be ground.

--

John Larkin, President
Highland Technology, Inc

jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro   acquisition and simulation

is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

Strange, current mirrors works for us but we don't use that configuration as shown here. A environment differences may also be a factor.

jamie

is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

Some people really like 2-transistor mirrors. That works OK inside of ICs, where parts are really matched, but even then I suspect that precision is seldom required. Made with discretes, with diffferent c-e voltages, self-heating of non-isothermal transistors will wreck any accuracy.

Suppose theta was 300 K/W. Dissipate 10 milliwatts. That's 3 K temperature change. That corresponds to around 7 millivolts of Vbe change. That's mumble mumble a lot of current error.

Simple mirrors invert signals just fine, but doing it to 1 or 2% accuracy is hard.

--

John Larkin, President
Highland Technology, Inc

jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro   acquisition and simulation

is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

Do the math on discrete devices with "ballast" resistors in the emitters... 200mV of drop, or more, dramatically improves matching of currents. I think I wrote that up somewhere on my website... I'd look it up and provide a link, but it's wine time right now ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

Ok, but I try to keep all of the discretes in close approximation of each other and loads even on both sides of the mirror.

One of the applications I use it for is to pass a signal from a sensing source that I do not want to have direct connection to. Sort of an isolation between sensor and output. All I need to worry about is 2 Mev dosimetry just under the enclosure that is housing it! :)

Jamie

ly. It is more compact than a diff amp or current source with inductor, but= there's no cost savings with DK's pricing. It is standard current source l= evel shift except it is voltage controlled by the 3.3V so the shift tracks = with power supply drift.

Huh? These should be biased somewhere in the 100u or less range which puts = Pd deviations in the .1 mw or less range for 0.03oK or 0.06mV Vbe or less, =

0.03mV deviation, which makes current ratio errors +/-1% range. And it is n= ot the difference in supplies that determines currents since the first PNP = emitter is approximately GND, it is only the 3.3V to GND potential that det= ermines the current.

It is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

deviations in the .1 mw or less range for 0.03oK or 0.06mV Vbe or less, 0.03mV deviation, which makes current ratio errors +/-1% range.

Do you think that part can hold 1% as a current mirror? It's not monolithic.

first PNP emitter is approximately GND, it is only the 3.3V to GND potential that determines the current.

Oh, OK, but it's looking at the 3.3 volt supply, not the EL output level. The advantage of looking at both Q and Qbar is that doing that takes out the chip's signal drop down from Vcc, which has a 300 mV spread in the specs.

The MAT04 (monolithic quad NPN) data sheet claims 1% mirror accuracy with the 4-transistor mirror topology.

--

John Larkin, President       Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

satisfactorily. It is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

Pd deviations in the .1 mw or less range for 0.03oK or 0.06mV Vbe or less,

0.03mV deviation, which makes current ratio errors +/-1% range.

problems anyway.

I'd like to keep the error below, say, 0.1 volts. The pecl swing of

0.8 volts p-p at Qbar is just a bit more than the fet needs to go solid on and off. My diffamp circuit only sees the Q:Qbar differential swing and ignores the common-mode voltage almost entirely... as well as the resistors are matched.

on that pHEMT then a tried and true expedient would be a wide-band RF autotransformer to boost it up to 1.6Vpp.

It's got to be DC coupled, which is why it's such a hassle.

developed a line for single ended to differential conversion for high speed A/D converter inputs about 15 years ago, weren't real pricey AIRC.

first PNP emitter is approximately GND, it is only the 3.3V to GND potential that determines the current.

cheaper to make that way. Shouldn't the pairs be monolithic at least? That's all you really need.

The Fairchild data sheet mentions "4 die".

Mirrors are intellectually appealing, but I doubt that they are accurate in real life, especially the minimal 2-transistor non-monolithic kind. They sure Spice nice.

--

John Larkin, President       Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

cheaper to make that way. Shouldn't the pairs be monolithic at least? That's all you really need.

!2die4

Siliconix HFA3096BZ are on a single die with dielectric isolation, but still the transistors are not pairwise interleaved as we see it in better op amp input stages.

Gerhard

ps

Any idea for a current source that can be switched on/off very fast? Absolute current +- 20% is ok, but must be stable. I currently use Wilson current mirrors in a time stretcher, and the ringing limits the linearity. (12-22 ns ---> 12-22 us or so, dual slope) Every ps counts, must be space proof, and anything more newfangled than a 2n2222 is frowned upon (rad tests etc..)

cheaper to make that way. Shouldn't the pairs be monolithic at least? That's all you really need.

I have used differential ecl to drive a pnp differential pair, in current-steering mode. If you use fast transistors, it may need base resistors to avoid oscillation. That steered current could drive a resistor on the gate of my phemt. Figure that would add a nanosecond to my signal path, maybe a bit less. "Every picosecond counts" for us, too.

Here's a thing I did once, for the NIF timing system. It worked, but it's a bit too complex, in retrospect.

The comparator is a hack. I designed it to use a MAX9690, but Maxim stopped shipping and a couple of months later announced that they would never ship. So I had to hack in the 9691's with diodes. We also had to replace about 3000 flakey 9690s in the field.

--

John Larkin, President       Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

This link has been updated with the emitter-driven version...

Eliminating some pole-zero compensation (as predicted). ...Jim Thompson

-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Why the 5K base resistors? They can make a lot of error.

--

John Larkin, President       Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

You don't know? ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

is more compact than a diff amp or current source with inductor, but there's no cost savings with DK's pricing. It is standard current source level shift except it is voltage controlled by the 3.3V so the shift tracks with power supply drift.

Reposted with emitter-driven added to the PDF...

Then there's the mirror way... not quite as precise as with the OpAmp (virtual grounds not as stiff), but not too shabby...

You used to be able to get 2xPNP plus 2xNPN in one package, such as TPQ6502, but I don't know if they're available in the current "teeny" packages.

(Note that _none_ of my methods used VCC as a reference... everything is referred to the actual PECL common-mode.) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.