Anyone know a source for cheap resistors (0402 or 0603 package) that are usefull up to around 4GHz? Small quantities ofcourse. The ones I found so far costs around $7 each.
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... "If it doesn't fit, use a bigger hammer!"
--------------------------------------------------------------
Since you are in Europe you could ask here:
Or ask these guys where the distributors are:
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Mouser. Digikey. Ordinary one-cent surface mount resistors work fine at 4 GHz. At very low impedances, it's sometimes worth putting two in parallel, like for terminating a 25 ohm line.
At higher impedances, the pad parasitic capacitance starts to dominate, no matter how much the resistor costs.
What are you working on? How precise do the impedances need to be?
If you're working with stuff like this, a great investment is a 20 GHz TDR scope, so you can fool around with parts and layout geometries and such. It doesn't take many board spins to pay for an ebay TDR system.
John
This was in the 90's, not sure how they behave today: If the meander cuts for trim were grossly non-symmetrical it helped to use two resistors of twice the needed value on top of each other, with one being upside down. Probably best to flip the lower one so you can still read a part value.
If they just weren't so freaking bulky ...
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Never tried at 4GHz, however it works for sure at 2.4GHz.
IIRC the optimal SMD resistor value for RF is ~100R, as it behaves resistive in the widest band. They don't recommend using parts over ~250R and below ~50R.
Vladimir Vassilevsky DSP and Mixed Signal Design Consultant
We recently did a sort of fast pulser
and tried mounting some of the 0603 resistors upside-down. It didn't make any difference. This box can make 50 ps edges, which corresponds to about 7 GHz bandwidth, all surface-mount Digikey-type Rs and Cs on FR4.
I sometimes splay out two resistors at 90 or 180 degrees, as terminators, to reduce stray inductance, but still flat, not stacked. Seems to help a little.
I hire big, young interns.
I'd love to do a tiny USB TDR pod, but it's hard to compete with all that stuff on ebay.
John
Nice enclosure. Who buys this kind of gear? University research labs?
I meant two stacked and one of them flipped. But possibly the deposition process is more precise than in the 90's and the meander cuts are miniscule nowadays, then it wouldn't make a difference.
Main thing is, you've got the space. Plus that crane :-)
But from what you said a while ago most of the sampling heads of the TDR scopes sold 2nd hand are shot. I can't imagine the ordinary user figuring out how to mess with the diodes and stuff in there. Of course, who needs TDR this fast and accurate? Installers narrow it down to a few feet and then fire up the backhoe.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
That's because they don't have 1 mm accuracy. ;)
11801/2 scopes are absolutely beautiful. The 11802/CSA803 version has a couple of nice coax delay lines replacing two of the four sampling plugin slots, which is usually a win because you really need the trigger delay and most of the plugins are duals anyway.In my IBM days I occasionally used all four slots, because one of them was taken up by an optical-electrical converter (i.e. a photodiode in a box) and I had one of the 50 GHz single channel plugins. But usually the convenience of the delay lines is worth swapping out the pods occasionally. Now if I can just score one of the 2m extender cables...you can't run coax to the sampler at 20 GHz, you have to hang the board off the plugin or use an extender cable. When my ship comes in, I'll go look for a nice SD14 with all the little doohickeys--3 GHz, dual FET probes--100k//0.5 pF. Dead useful.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal ElectroOptical Innovations 55 Orchard Rd Briarcliff Manor NY 10510 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
An update. I was reminded to go through my lab wants list and do a bit of maintenance on it, because my $350 HP 35665A finally arrived after being stuck in a customs warehouse in Jersey for over 2 weeks. (A bit of shining up and a twist of one pot, and the display is bright enough to fry eggs on. Beautiful.)
Sooo, I flipped over to eBay and looked for SD14s. Whadda ya know, there was one going for cheap, and *ending in 5 minutes.* So now I'm waiting for my new $175 SD14 (hopefully it'll actually work).
This year stinks in some ways, but it's an amazing time to be buying test equipment.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal ElectroOptical Innovations 55 Orchard Rd Briarcliff Manor NY 10510 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
We did the gaussian pulse version for a Canadian laser company. We put in the hooks for the other versions and put it up on the web as a standard product. I did the pcb layout myself! We'll get some press releases out of it, too, and maybe somebody else will buy some.
We may do a 6-volt version too, to drive e/o modulators directly.
And the water-powered hydraulic freight elevator!
Nearly all the ebay claimed-good heads have actually been good, or the sellers took them back. A few of the "untested" ones had a channel out, or were totally dead, but they were cheap.
Here's the insides of one...
ftp://jjlarkin.lmi.net/SD20.zip
As you can see, the guts are unrepairable. Sometimes the eeprom dies, and the scope won't talk to the head, and I suspect I can replace that eeprom with one copied from a good head. I have maybe a dozen like that. Spare-time project.
I don't think we've ever blown up a sampling head in-house. We use 20 dB attenuators whenever possible.
John
Maybe you could let a company like Thorlabs carry and sell it.
Wow, maybe you can put that on your web site, into a "history" section.
Amazing. I wonder where they got the super-small SMT parts for the actual head, considering that these were probably built in the 80's. Although picture 005x seems to show a 1995 number.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Congrats!
True. Same for industrial real estate. Except that out here one never knows whether and when the left blows a hole into the business property tax cap (Prop 13) which can cause an exodus. Plus a lot of places have been vacant for over a year so they are all negative cash flow.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
These heads were hybrids, raw, unpackaged silicon chips wire-bonded to a metalized ceramic substrate. That way predates what we call surface mount.
John
Ok, that sure would make them unrepairable for ordinary folks. Maybe occasionally you could find a SMT diode or something that's small enough to be crammed in.
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
I'm looking around for some lab/office space myself, just on spec. There's lots around. Still, the taxes are quite uncertain round here too...leases appear always to include only the base year's taxes. The competing possibilities are to move house or add on...ick.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal ElectroOptical Innovations 55 Orchard Rd Briarcliff Manor NY 10510 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
At decent signal levels, the SD14 probe doesn't need a ground clip. There's so little input current that waveforms look fine without a ground, just a bit more fuzz from EMI. With digital stuff, you can do useful probing just getting close to the signal, not actually touching it at all. You can spot oscillations by just waving it around a board, free-running trigger, and look for "noise." The hot probes do make your hands sweat.
Great gadget.
The SD-series extender cables are hard to come by, and cost $500 or so on ebay. But they are mighty handy.
John
But you've got a fairly big house. If the kids move out to wherever their universities are you'd have plenty of room. Also, check the rules about granny flats. In many jurisdictions it's pretty easy to put one up if not larger than 1200sqft or so. You'd have to inquire first.
Then, get a license platr frame like a friend of ours has, "My daughter and my money go to University of ..."
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
It's about 2200 square feet, with 5 larger-than-life personalities in residence at the moment. Our #1 daughter Bronwen graduated a couple of years ago, but is off work with a badly herniated disc, and she's stuck here because there's a fight with the Workers Comp folks that means that her health insurance won't pay for surgery until the comp claim is settled...#2 daughter Magdalen is taking a year off university, working part time and doing some music classes. She has a wonderful opera voice and likes singing a lot, which makes the house sound better but seem smaller. (She sounds a great deal like Audra Macdonald, for fans of that sort of thing.) #1 son Simon is a senior in high school, is nearly 6 feet 6, and would like to have someplace to bring his friends and to watch football. So Maureen and I are hip deep in offspring at present--which is really rather pleasant, but definitely crowded.
Thus my taking up space in the basement and needing lots of quiet doesn't fit too well with the rhythm of the place, unfortunately.
Yup.
Cheers,
Phil Hobbs
-- Dr Philip C D Hobbs Principal ElectroOptical Innovations 55 Orchard Rd Briarcliff Manor NY 10510 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
Sounds like I can get away with that too. I'll probably use 0402 parts to preserve board space.
Its an idea for a measurement product (don't worry, nowhere near the stuff you are doing). I need the resistors mostly for transmission line termination. Since traces on FR4 and the internal impedances of the chips aren't very accurate 1% precision will do.
I have been looking into a sampling scope but they are bulky (that will cost me at least $200 for shipping) and ofcourse the heads may be bad. It does take a while before a complete kit for a reasonable price comes along.
BTW for some unknown reason that sort of 'special' equipment is very very rare on the 2nd hand market over here. The last two logic analyzers that I bought and most of the accesoires came from the US.
--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
"If it doesn\'t fit, use a bigger hammer!"
--------------------------------------------------------------
Been there, done that. Now the two of us have 3650 sq.ft. to ourselves, except on holidays, when there's likely to be 20+ for dinner ;-)
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
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.