Large capacitance varicaps, where are thee?

I receive all ON product notices (spec, process, material, source, discontinuance, resurection etc etc etc) changes, and there's nothing on SVC383.

Did you ask the ON recommended vendor - Rochester Electronics? They also have no stock, but they shouldn't give you a phoney reason for this. Digikey still lists their supplier for SVC389 as Sanyo Semiconductor (USA), so their info for both parts may be dated. Did you actually ask Digikey? Looks like the usual minimum purchase of a reel, for items not recently stocked.

The issue is more likely a Sanyo-based supply chain one - strictly sales related. I'd sniff around previously reliable Sanyo distribution outlets (not excluding Digikey). SVC389 was also an RS Component line item.

RL

Is this pin where you see the sawtooth? Seems to me with a high impedance current source, you can steal the current. Now a triangle waveform is a different case.

So you have tried a high impedance current source?

Single source because you want a crappy behavioral model to make you feel good. There *are* other vendors out there, with perfectly good products.

Handle it.

You think you're alone? Get the manufacturer to guarantee the design.

Have you looked at Big Fat Schottkies? The 30V 40A types are in the 10nF at zero bias, and obviously will handle the puny 50mA reset pulse. :)

Can't say I ever got great results using schottky capacitance, I think their Q is crap or something (which is something to think about when everyone touts "zero reverse recovery losses"!). It also varies rather quickly with voltage, maybe too rapidly for the chip. Well, you can bracket the C range with series/parallel caps anyway, and maybe a higher voltage schottky, or a beefy junction diode, will have a more gentle curve.

Tim

-- Deep Friar: a very philosophical monk. Website:

On a sunny day (Mon, 22 Oct 2012 14:22:49 -0700) it happened Joerg wrote in :

They are upstairs in the attic here.

Joerg a écrit :

Medium size Mosfets?

Reading between the lines on the LTC3721-1 datasheet, it seems that the CT pin generates the upward part of the timing ramp with a (roughly)

In an earlier post, Miso suggested sinking some current out of the CT pin to reduce the frequency, to which you replied "... the chip immediately gets sea-sick when you do anything DC to that pin."

Had you tried *sourcing* a small (a few uA) current into CT to increase the frequency?

Regards, Allan

Regards,

Boris Mohar

Got Knock? - see: Viatrack Printed Circuit Designs (among other things)

void _-void-_ in the obvious place

I tried simulating LTC3721 and the charge waveform looks a bit funny, it is ~160uA but at the end of each cycle there's a different ramp upto somtimes -25mA before the 50mA reset pulse

adding some current, 10V - 50k, ~doubles the frequency and does not seem to make it go crazy

-Lasse

The whole tuning range of the hyperabrupt parts was achieved by a 6-8V span of bias. If the ramp amplitude is a signifigant percentage of this, you risk rectification and other effects.

RL

On Mon, 22 Oct 2012 14:22:49 -0700 Joerg wrote in Message id: :

How about

I can't go below 3V because that's my working voltage, with margin. So realistically I'd get 10-60pF range, times two. Not bad at all. I am just concerned that it might go unobtanium like all the others. Much of the BB series already is, and some places like Allied are cleaned out of this one right now.

The solution I have now is a mux plus 1% caps. I was surprised how cheap they've become.

The capacitance will modulate which will cause slight non-linearity. Would be acceptable in this case. My worry is the supply chain, not the technical side.

Yup. The chip then starts to misfire, big time, at least on the simulator.

Real estate. Those things are huge, plus they barrier is quite fuzzy.

This one has to be fast design and fast turnkey. But there was some good old bench time, with their old system so I can figure how the load reacts and stuff.

But no simulator models, plus they would also be single-sourced. Some don't have current-mode control and that's a no-no here. So far pretty much all my designs came out almost verbatim as simulated. That's the kind of time-to-market clients want.

No, they list ON Semi as the supplier.

FindChips lists only Digikey as the supplier and they state "obsolete" (on the ON brand, not Sanyo). No other distributors listed, and I tend to go by the more major ones.

ON Semi's tuning diodes tab does not list the SVC383:

Keyed it into the search field and it didn't find it.

That's too much "writing on the wall" for me to use a part, no matter what a sales guy would tell me over the phone.

Tried it, the results were less than enthusing.