Ditch your power plane and star route Vcc?

New brain ticking along nicely?

Cheers

Phil Hobbs

Reply to
Phil Hobbs
Loading thread data ...

? Maybe they were stuck in a publish-or-perish hole?

Reply to
Robert Baer

NymNuts, Still trapped in the body of a fairy, still not escaping the filter ;-)

...Jim Thompson

-- | James E.Thompson, P.E. | 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 | Stormy on the East Coast today... due to Bush's failed policies.

Reply to
Jim Thompson

1.pdf
B

one of

o the

mesuch.

I know that RC decoupling is very viable in the vast majority of applications, but in the very high gain broad bandwidth applications I know of all you're doing is defining the phase shift and frequency of oscillation :-).

HP had a series of app notes showing how to solve many such problems in sub-microvolt-level instrumentation with star grounding and power distribution... I think 60's and 70's era.

Tim.

Tim.

Reply to
Tim Shoppa

somesuch.

Sure, you can't just slap down 10ohms and 0.01uF, dust off your hands and walk away :-)

Back then you could often get by with it but when devices still have considerable gain at 5GHz or 10GHz a star ground will be a huge challenge. Not so much in terms of stability but noise.

--
Regards, Joerg

http://www.analogconsultants.com/

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

of

somesuch.

This is about 120 GHz of gbw in a box about 3" long:

formatting link

I scribbled out the schematic for this as a simple PCB layout exercize for The Brat, and it worked, so we put it up for sale. Things like this generate press releases if nothing else.

Power is via big L3 pours, with ferrite beads and bypass caps picked off at individual amplifier stage power pins. L2 is ground.

With tiny surface-mount parts, this sort of thing isn't really hard.

John

Reply to
John Larkin

article:

formatting link

of

somesuch.

Maybe try to offer it to companies such as Thorlabs? Then you could spend the net proceeds at local businesses like Zeitgeist :-)

Did you employ sizeable L3 quiet pours at the individual stages or just big enough to via the bypass caps there?

SMT has made a lot of things easier, like clean transitions into controlled impedance traces without the usual impedance bump at the pin. I was lucky in that my first employer was a very early adopter. All my designs there (and those of all others) were SMT. Huge boards. Initially we had to run our own reflow ovens because there weren't many assembly houses that would accept 3HE size boards. Intense cooperation with the production folks there was crucial to avoid problems like warpage and being only five minutes walking distance away helped a lot. They just didn't like it when we came after lunch, having had Greek gyros with tzatziki and tons of garlic in there.

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

I regularly partition supply rails, depending upon local decoupling for performance. Switching currents are nailed locally, whether the supply distribution is poured, or not.

Belts and braces.

RL

Reply to
legg

Hi John,

There does have to be a return current, it's just that it can be a displacement current rather than anything requiring copper.

We use plenty of Mini-Circuits-type MMIC amplifiers (e.g., Gali-3), and one design had 50 ohm line->parallel caps of something like 1nF and 1uF->amp. My boss was mentioning yesterday that he had a tech Dremel away the pads for the second cap so it's just 50 ohm line->1uF->amp and gained himself 2dB less ripple across the amp's response (something like 100kHz-3GHz): The parasitics/impedance bump from the second capacitor pad do a lot more harm than what might be frequencies getting close to the self-resonant frequency of the 1uF cap.

---Joel

Reply to
Joel Koltner

Careful, if you get this much change in amplifier behavior by removing a high-frequency style cap then something might be riding on the edge.

--
Regards, Joerg

http://www.analogconsultants.com/

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

That's interesting. One of my gadgets has a strange - not harmful, just weird - bump in its frequency response, around 3 GHz, well into its rolloff region. Looks goofy. Maybe I should play with bypass caps.

The SRF of the 1u cap will be quite low. 1u and 1nH resonates at 5 MHz. But there's nothing wrong with using a cap above its SRF. A 1 farad 0603 cap would self-resonate at 5 KHz. Hand me that Mouser catalog, please.

John

Reply to
John Larkin

article:

formatting link

one of

somesuch.

There's actually a small L1 mini-pour at each power pin, so the bypass caps can get a grip.

ftp://jjlarkin.lmi.net/J750.gif

Once while cooking I said out loud "I wonder if I used too much garlic" and my wife responded "excuse me, I can't understand those words."

John

Reply to
John Larkin

article:

formatting link

one of

the

somesuch.

Those are small, not really pours but brush strokes :-)

Same here. And same with spices. Whenever we host visitors from Europe we have to make sure to tune it down a few notches with the hot sauces and jalapenos.

Our local Thai place changed owners and somehow hot ain't that hot anymore. Next time I am tempted to order Thai-hot, usually only tolerable by natives from there.

--
Regards, Joerg

http://www.analogconsultants.com/

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

article:

formatting link

one of

the

somesuch.

Genuine Thai food isn't flavor, it's all pain.

Do you know about Paul Prudhomme's Redfish Magic?

It's great on catfish or talapia.

John

Reply to
John Larkin

off

Sure. Don't do things that won't work.

John

Reply to
John Larkin

#chuckle#

Three other chili-heads and I from my office went down to our favorite local Thai place a couple of years ago. We ordered our favorite lunch specials, and three of us asked for it Thai-hot. As usual, the waitress asked us for confirmation, twice (it's apparently a Hunting Of The Snark "What I tell you three times is true" protocol).

We could tell when our lunch orders got to the head of the line in the kitchen. We heard coughing... a wave of it, starting from the tables nearest the kitchen and moving across the room in our direction. When the wave reached us, our eyes started to burn and *we* coughed. Lots of hot pepper in the wok...

When served, the lunches were definitely Thai-hot... the spicy hot catfish was memorable. It hurt so good!

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!
Reply to
Dave Platt

Not here. Have to watch the place under the new ownership but so far they've kept the cook.

Thanks, have to try that. We love tilapia.

--
Regards, Joerg

http://www.analogconsultants.com/

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

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.