X60008 Voltage Reference IC

[floobydust]

Whatever happened to the X60008 ?

It didn't incorporate a heater, accuracy ±500?V (±0.01%) initial, max. 1ppm/°C, aging 10ppm/1khrs and uses XFET.
It looked pretty good, I have some in a Malone Electronics low cost voltage reference. Nut's n Volts magazine build 9/2007

That semiconductor merger & acquisition crap comes along. Xicor originally designed it 2004, then Intersil buys Xicor for $0.52 Billion. 2017 Renesas buys Intersil for $3 Billion (but Maxim wanted to buy them) and the IC's are gone now, Intersil has discontinued their highest performing references, odd politics.

Maxim has a voltage reference lineup MAX6335, 6241, 6250 that is similar in cost and spec, but I'm not a fan of their parts.

I can't see anything to replace it for the price point ~$7. Any suggestions?

Xicor/Intersil Voltage Reference Application and Design Note AN177

[Andreas]

It didn't incorporate a heater, accuracy ±500?V (±0.01%) initial, max. 1ppm/°C, aging 10ppm/1khrs and uses XFET.

It looked pretty good, I have some in a Malone Electronics low cost voltage reference.

Hello,

sorry but: its not XFET but FGA-technology (storage of the output voltage in a kind of  EEPROM cell).
I cannot speak for the X60008 since it was out of production when I started to examine 5V references.

Instead I used the ISL21009BFB850.
there are 3 issues which I have with the ISL21009BFB850
- switch off hysteresis
  when running for several days and then you measure the steady state output voltage.
  switch off for some seconds
  switch on again
  the ISL21009BFB850 starts some 100uV (around 200-300uV) below the steady state output voltage
  it needs some days until the steady state output voltage is reached again.
  attached a around 10 hours plot with 1 minute switchoff at the beginning.

  would be interesting if this "feature" is also available on the X60008.

- radiation sensitivity
  I have never seen a application note on any other reference to convince customers that a reference is enough radiation hardened.

- sensitivity to capacitive loads.
  you need a boucherot cirquit to stabilize the output when loading with several nF capacity.

I would go for a AD586 or LT1236 for a really good device.

with best regards

Andreas

[floobydust]

Thanks. I mixed that up, it's FGA (floating-gate analog) technology I was not familiar with.
"...Typically the trapped charge on the floating gate can remain without loss for greater than 10 years."

Hmmm a bit skeptical on the long term stability.

NASA took the X60008, ISL60002 down to cryogenic -195°C temperatures. Measured drift was 3.25 ppm/°C

I'm sure radiation would knock out those stored electrons, odd NASA would even consider the technology for space exploration.

[montemcguire]

Intersil has an application note that addresses this issue: http://www.intersil.com/content/dam/Intersil/documents/an15/an1533.pdf

Yes, it is sort of creepy, but they say that a relatively thin zinc shield can filter out the X rays that cause problems.

An odd possibility: the FGA voltage could be trimmed by exposure to X rays post-assembly. Probably only trimmed in one direction, but heck, it seems possible. :-)