Is National getting out of the switched capacitor filter business?

Whosever you use, be careful about aliasing the signal, aliasing any power supply noise, input charge-injection spikes, and output spikes!

And noise. They tend to be noisy.

John

S/C filters have never really caught on, IMHO because they wanted to cram a pricing structure into the market that made these filters unattractive. I am probably one guilty party, having used them on only one design where I had to but not on any others. The sole reason was the high pricing.

Anyhow, personally I trust LTC a lot. The earth would almost have to be hit by a huge meteorite before they discontinue something. Maxim, ahm, well, you've already said all there is to say. Even if they have parts there is never a guarantee that they'll have parts next time the purchasing folks want to order a reel ;-)

Other source, and this might actually save money on some products: Cypress. Many of their PSoC contain several S/C filters, along with other useful analog stuff.

John Larkin writes

Aha. So it's not just my lousy layout. I only tried using them once, and was stunned by the (power rail) noise. A simple 100nF decoupler to +5V and -5V just didn't cut the mustard.

However, I generally like and trust Linear Tech stuff. The NatSemi filters didn't seem to have devices with such high specs in their lineup, I got the impression they'd perhaps come up with the idea in the first place but Linear Tech had leapfrogged them and NatSemi had stopped developing them years ago; if NatSemi find they are not selling, they may have stopped emphasising them in their main index?

You've listed all the SC manufacturers I know of.

Can't confirm that. My one and only design with lots of S/C filters on board was a medical ultrasound Doppler board. There you need in excess of 80dB in dynamic range, and it delivered. But you've go to have at least partial planes for the supply.

I think I will be changing the whole topology but what I did was: The input of the filter is fed with a fairly simple RC filter from something that should be down by 60db (in theory) by time we get to Nyquist.

The supply is from a linear regulator.

The input side spikes should all end up in the capacitor. The output side fed another simple RC to kill off the steps.

My signal is over 2Vp-p and I only need the noise in the

I'd really like to start trusting Cypress again. I will look at the PSoC chips again. Even though I am changing topology I may find something worth a look.

This may even end up as an ASIC. Currently, the entire hand held device will fit in the back of a pick-up truck.

Where did they disappoint you in the past?

Oh, the Silverado edition of your design :-)

Bwahahahaha!

Reminds me of the first mockup of an electronic transmission controller at Ford (~1968)... the "breadboard" filled the trunk of the car :-) ...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

They quite making a part just after I designed it in. I couldn't really get too made at them because it was a CPLD. At this point

Do they come with a 2nd battery option? I have met the design goal that it run on a single battery. A 100 AH deep cycle isn't quite what marketing had in mind but they are just a bunch of whiners.

Marketing didn't like my proposed name. I wanted to call it the Hindenburg-Titanic Jr and give it the part number

Programmables are notorious for vanishing. That's why I avoid them as far as possible. Once I almost designed in the Intel series. Man was I glad that I didn't because two years later ... poof.

No idea (probably not as a factory option) but the guys with campers on there usually have an additional deep-cycle.

AFAIK the abbreviation "SMT" sounds like death in Czech, I believe it's "smrt" in their language. They are pretty frugal when it comes to vowels :-)

It is that way in Korean. For that reason, Koreans think of number 4 as a symbol of bad luck.

VLV

I have heard tell that they regard "R" as a vowel.

-- "Electricity is of two kinds, positive and negative. The difference is, I presume, that one comes a little more expensive, but is more durable; the other is a cheaper thing, but the moths get into it." (Stephen Leacock)

On Mar 20, 6:02=A0pm, Joerg wrote: [....]

Things like PICs and 8051s seem to stay around for a while. CPLDs seem to only go away if I design them in. What was the Intel part IIRC a while back I did design something from them into a product too.

I have been thinking that perhaps I should charge companies to not design in their parts.

In my latest design, I have several micros and no CPLDs or even

I am using a lot of 74HCxx logic in the design in an effort to keep the power low. One part needs an oven so there are limits on what I can do with circuit design. I really kind of wish that the LT1352's supply current didn't rise so much when you slew them hard. I have had to reduce amplitudes and move to a slower op-amp in one place.

Seriously: Some folks want to go with some 3 AH Lithiums. Do you have experience with this technology. It worries me that just a little abuse can turn a Lithium into a rocket motor.