1. Home
  2. Electronics Design
  3. Bidirectional logic level shifting from positive to negative supply
  4. Page 2

Bidirectional logic level shifting from positive to negative supply

formatting link

It is a local client so you can get there by car? Sometimes I wish I had a pilot's license.

Lower threshold comes with speed penalties. You'd have to pick one that represents a good compromise.

Of course, then there is the other trick: NRZ magnetic transmission. Essentially a glorified Schmitt trigger with a li'l signal transformer up front. A comparator may be best for that because there you can set the hysteresis, the open collector output is included in the deal, and you can get almost any speed you want. The downside is that the first bit after turning on the circuit is undefined because the center-biased Schmitt will wake up randomly. Can be goosed by the POR/BOR though, which any respectable uC should have anyhow.

I'd use one of the Texas iso-converters. Avoids the hassle of "Oh, someone must have left it on overnight and now the battery is dead".

That's a low stock boutique part, probably have to hurry before they are gone. But 5.5GHz? Yikes. What speed do you really need?

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
Reply to
Joerg
Loading thread data ...

Interesting, thanks. I wouldn't have thought it was that low.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply to
Phil Hobbs

formatting link

Nope, Albuquerque NM. I'm there now.

Yup, probably right.

If I just slow down the processor for debug--which would be fine in this case, it'll be loafing anyway--1Mb/s is probably OK.

I'm actually using the HFAs in a new-improved laser noise canceller design, where I need the matching. They're pretty nice parts.

The main reason for my paranoia about this is that I haven't had any experience with the DebugWire interface, and I do _not_ want another board turn.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply to
Phil Hobbs

Keep in mind that these things spew around a fair amount of EMI. So if there is noise sensitive stuff on either board you might want to add circuitry to gate off the DC/DC converter.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
Reply to
Joerg

formatting link

Probably be getting some nice thunderstorms tonight.

[...]

If you are there a few more days you could fire off a Digikey order right now, get the stuff sent pronto and lash it all up with rework wire. IBM-blue rework wire, of course :-)

Then you'd know that it works.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
Reply to
Joerg

[snip]
[snip]

Probably already got them. We had a "gully-washer" here last night. ...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

formatting link
| 1962 |

Spice is like a sports car... Performance only as good as the person behind the wheel.

Reply to
Jim Thompson

Yes, this seems the right approach. So, make your (+) current mirror to feed the cathode of the zener, the (-) current mirror to feed the anode of the zener, and the two mirrors are programmed by a resistor that connects their inputs (for 10 mA through the zener, that'd be 860 ohms).

Wasn't that the original schematic? I get confused about the biasing in that one.

Reply to
whit3rd

Proper current mirrors don't need to be in isothermal environments.

It's still hard to find mirror-quality monolithic matched dual transistors. Analog Devices discontinued most of theirs, and what's left is insanely expensive. The common "matched" surface-mount duals are terrible; two chips on separate lead frames, terrible thermal coupling. Here's a thermal image of a UPA800 with 80 milliwatts dumped into one side:

ftp://jjlarkin.lmi.net/UPA800_80mW_one-side.jpg

pretty much useless as a classic current mirror.

They don't need to be in ovens, either.

John

Reply to
John Larkin

Pretty much. The trick is to get the currents to match closely enough that you don't net leak much out into the microprocessor/jtag stuff.

There should be some tricky opamp/diffamp circuit that emulates a floating battery, but I haven't had my coffee yet.

There is some old ideal-transformer model that uses cv and cc sources. I used to use that in ECA, an old DOS simulation program.

John

Reply to
John Larkin

Is there 'anything' you can't blow up? ;-)

Reply to
Michael A. Terrell

You wouldn't need one, if you'd just keep your gullies clean. ;-)

Reply to
Michael A. Terrell

If we took away the words *POOF* and *PHUT* Joerg would become speechless :-) ...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

formatting link
| 1962 |

Spice is like a sports car... Performance only as good as the person behind the wheel.

Reply to
Jim Thompson

The mirror at the left is just there to turn on the pullup when the +5V goes away (i.e. the programmer is disconnected). As soon as the +5 is attached, it turns off and the programmer takes over. Only one mirror is working at a time.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply to
Phil Hobbs

It wasn't me who dunnit :-)

No, only the failure modes would change. To *KABLOUIE* and *BANG* ...

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
Reply to
Joerg

Hey, Phil,

Does this work?

ftp://jjlarkin.lmi.net/Level_Shifter.JPG

The comparators, at least the lower one, will need separate V- and output common pins. There's probably some way around that, too.

John

Reply to
John Larkin

IN- of the lower one might go *PHUT* :-)

They usually can't take more than -600mV versus the negative rail.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
Reply to
Joerg

I mentioned that gotcha.

John

Reply to
John Larkin

Hmm, I like the idea.

The 200 mV low state offset prevents the line from latching permanently the first time it goes low, check. Something like an LM319 would probably work, if there are no funnies when the +5 goes away. The 319 doesn't go to the rails, so the -4.8V threshold is a bit of a puzzle. I have an unregulated -9V supply I can use for that. At 80ns or so, it'll be quite a bit faster than my version, and at a Joerg-esque price.

I'll fill in the resistor values and figure out how to make the +5 go away gracefully, but we may have a winner, thanks.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply to
Phil Hobbs

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

I wonder if that looks any different if the collectors are tied with a big pad of 1 or 2oz copper (not that such connection is useful for mirrors...).

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

The LM319 can run off +5 and -9, because the emitter of the output transistor can be anywhere within the supplies. We'll always be within the CM range that way. Still have to figure out what happens when the positive supply goes away.

I gotta start thinking of comparators sooner--overlooking this sort of thing has happened to me a few times now.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply to
Phil Hobbs

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.