Maxim Semi Deliveries

You do a post and you don't give the part number?

Do I have to? :-) To satisfy your curiosity: its the MAX4447

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

Oh, right.

It wouldn't have quite accurately half the gain you expected? Like within 2% or so?

Jeroen Belleman

(the Arcor news server is unwilling to give me Nico's post)

Ouch! Finally, after 13 years, someone discovers this.

Have you seen that the part is single-ended-to-differential and that Vout is clearly described as Vout+ - Vout- under "conditions"? Some people DO have differential probes.

regards, Gerhard

There's a reason for the "Incoming Inspection Dept"

I like the LTZ1000 datasheet:

"For applications not requiring the extreme precision or the low noise of the LTZ1000, Linear Technology makes a broad line of voltage references. [...] Only applications requiring the very low noise or low drift with time of the LTZ1000 should use this device"

It's like, it would be an *insult* to use this part in an application that does not deserve it!

--

John Devereux

I clearly recall it has half the specified gain. It has been years since then so don't ask me about the exact details :-)

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

Maybe they don't want to deal with the support calls from people who can't get it to meet specifications.

True, it is an awkward beast. Still the king of references though, after all this time (decades).

--

John Devereux

Great part, though. I have about half a dozen in the drawer.

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
845-480-2058

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

Hmm, curious so I looked it up. References AN-82 and AN-86. AN-82 (Mitchell Lee, 1999) includes a thermal noise test: the board was covered with a foam cup and measured; the cup was then removed and the measurement continued. Noise jumps from a few uV to about 30uV. "Ambient in both cases was a lab bench-top with no excessive turbulence from air conditioners, opening/closing doors, foot traffic or _547 exhaust_." (my emphasis). A dig at Jim, perhaps? :) The second one, AN-86, was in part authored by Jim, as evidenced by the gentle sarcasm and mild technophobia (the 546 photos, and preference of toggling dials over surfing the internet).

Tim

-- Deep Friar: a very philos>

t

h cases

A dig

, as

d o r

not

an

Was the noise thermally generated or optically generated? I know when I worked down at the femtoAmps region, I had to turn off the lights and no one could walk around! After all a few electrons moving by, looks like current.

do

For

not

Or maybe it is an oblique hint that it is expensive.

?-)

It's thermal.

The LTZ1000 is very sensitive to air currents around it and the circuitry connected to it. Actually that's not quite right. The part

*itself* is stable, but its Kovar leads make thermocouple junctions with the copper PCB traces. It has an on-board heater to stabilise its chip temperature at some elevated value, 80 degrees C say. So there is a largish thermal gradient between the chip and the board. Any air currents produce local cooling variations so you get thermocouple voltages generated which don't always cancel precisely.

You also have to be careful not to allow an external thermal gradient to appear accross the device (for the same reason). Some people put cutouts around the part to prevent this. Do a google image search on "ltz1000 pcb" and you will see several examples. Look out for the pink foam insulation typically used to keep out air currents.

--

John Devereux

Me too :)

Well just four of them to be precise. In my hobby-mode I have a test circuit set up for it. I am learning a lot just figuring out exactly how the circuit works, making measurements, seeing where the errors come from.

One issue is measuring the drift. There isn't anything more accurate to compare it to! Nothing without pipes for the liquid helium anyway. The best meter is HP3458A which contains.... a LTZ1000.

--

John Devereux

Unfortunately, in a recent application, the thermal variations turned out to be right smack in the signal band.

do

r

or

not

can

w

How about if you set up three of them and measure the relative drift?

George H.

Well why do you think I have 4??? :)

Yes that is what you can do.

Here is one with 10

Assumes they don't drift the same way of course.

--

John Devereux

They should package one of these and all the support circuits into one of their cute little LTM modules.

--

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