Scoping Tiny Signals ~250uV

Ain't gonna happen. Use a spectrum analyzer.

Bummer :( I thought if I could get the signal big enough for the scope I could use the FFT feature built in. Or if there's a dominant signal, I could just look at the waveform.

He could buy and cascade a couple of MiniCircuits mmics-in-a-box, 20 dB each, but they're not very quantitative.

There is this:

I scribbled this out as a pcb layout training exercize for The Brat, and it worked pretty well, so we called it a product.

John

About the Highland amp.... Noise spec: 5mV RMS into 50ohm load. Gain@50ohm: 50X

So

Vout = Vin X Gain = 250uV X 50 = 12.5mV

I think I'm going to see 5mVRMS of noise added to 12.5mVRMS of signal(mostly 800Khz squarewave).

I don't think I'll be able to pick out what I want just by looking at the waveform. Butttt.. DSOs have averaging features :) So I might be able to see through the noise.

I might do a RFQ. Thanks

amp.

Why are there no prices on your web site? I hate that. Does your pricing compete with mini circuits? I've got a couple of their ZFL-500's which I think only cost ~$60 or so. (0.05 to 500 MHz.)

George H.

Marketing wants to harvest contact info and feed that to the reps. So stuff like manuals and pricing require filling out the form. I fought to keep the form minimal.

No, we don't compete with m-s, in products or pricing. The J750 is $750... I wonder how that happened.

John

an amp.

EMI

I keep arguing that if we put our manuals on line it will generate more sales. But no one listens to a rosin sniffing tech.

(Well a few of the manuals are pretty crappy, would they discourage sales?)

George H.

Haha... nice.

Shouldn't your specs there at least include some tolerances on the gain and/or plots of return loss, reverse isolation, etc. though?

I'd also be tempted to just through in the wall-wart power supply rather than having it be a separate purchase.

But overall it looks nice. We sell an "isolation amplifier" -- just amp/pad/amp/pad, providing >60dB reverse isolation up to something like

3GHz -- which is probably not as universally useful and I think we charge more.

Your DC offset comes from floating the entire amplifier chain as I recall, doesn't it?

---Joel

amp.

So Vxxx is for VME?

What do J, T, and P mean?

That would be in a 1 GHz bandwidth. The NF isn't bad, it's the bandwidth that piles up noise.

If you have a good trigger available, yes.

You'll get the SED discount!

John

We're time-domain people, so don't think in s-params. I can't even measure them.

Yeah. We're erratic about that.

Not here. Both input connectors are hard grounded. The zero pot is just a DC offset adjust, for nulling photodiode leakages and such.

It's really pretty simple, two opamps and some power supplies.

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

As I said, it was born as a layout exercize. My layout guy quit, and The Brat wanted to take over, so I needed something that could be done in the roughly 1 week overlap.

John

Broadband coupling transformer?

Probably your best bet, though, is to use a mixer to downconvert a small part of that range to a more convenient frequency with limited bandwidth, and amplify that smaller bandwidth.

Oh all right, I understand.

In the RF world the practice (including what we do) is to provide a feedthru capacitor as the power pin and a ground stud; that apparently gets you out of having to provide a power supply. :-)

Ah, gotcha -- I was thinking "MMIC amp" rather than "op-amp."

---Joel

amp.

The letter in a product name corresponds to a product line, and to the first two digits of a drawing number.

V = 22 = VME P = 23 = benchtop/rackmount instruments J = 21 = small boxes, mostly photonics T = 28 = bigger boxes M = 32 = CAMAC, mostly gone by now

and a bunch of others for OEMs and weird stuff.

For example,

V490 is a VME module.

22D490-A is the rev A pcb design 22S490-A is the schematic 22A490-2A is an actual product, dash 2 version, V490-2 to the customer 22A490.2A is the BOM (ascii MAX file) for above 22E492-B is rev B of the firmware 22C493-D is rev D of the FPGA

The BOM ties all the various bits together for a given deliverable.

John

Yeah, I hate having to solder the power supply connections on those things.

These are real opamps, which gets us DC coupling and relatively precise gains. MMICS are actually, usually, terrible from a quantitative standpoint: gain accuracy, actual impedances, frequency rolloff.

John

That's why.

RL

Cool! Thanks.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Me too. I suspect they think a lot of the alternatives wouldn't be as good when, in actuality, in many cases it probably just wouldn't matter. E.g., use an RCA connector for power (I'd suggest BNC or SMA -- and I have seen BNCs used for power on infrequent occasion -- but I wouldn't want Joerg to come after me for the extra cost involved :-) ), solder down some chip caps ASAP and OK, it's not quite as good as a capacitive feedthru, but probably hard to see the difference below a few GHz or so.

...and leave the solder-on terminals there for people who are just going to put it all in a bigger box anyway...

Ah, they should appeal to the "negative feedback is baaddd" audiophool crowd then. Hmm... that sounds like a good product -- a 3GHz headphone amplifier!

I imagine somebody somewhere has taken a MMIC amp and closed a loop around it.

---Joel

On a sunny day (Mon, 06 Dec 2010 13:10:50 -0800) it happened John Larkin wrote in :

The colors are nice :-)