air-wound inductor Q

Does anybody know how to permanently stop NASA Tech Briefs? I've told them many times to stop sending me that trash, and they do for a while, then they re-subscribe me.

I'll try having someone tell them that I'm dead. That might work.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

A t-t would work, if I can get the notch down around -50, or at least

-40. The cap values will be small, 300 pF ballpark, so there will be parasitics. There's not much that I've found, in books or online, about how to trim the null, or the effects of parasitics.

I could probably live with the passive version, if I calibrate the gain curve. The boosted twin-tee might be trickier, and opamp distortion has to be considered. I'll put some scut bunnies on the problem.

So, I need a 50-ohm lowpass or bandpass filter between my signal generator and the DUT, and a notch or highpass filter on the output side, ahead of a spectrum analyzer or scope FFT. The DUT output is differential, but some delicate hand-waving may convince the customer that one output will have more THD than the other, and that we can do a single-ended analysis on that one.

The photodiode itself will have 2nd harmonic distortion in the -40 to

-30 dB range, so it's overkill to spec our amp at -80. But someone did.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

I've given up worrying about that, since there will always be several companies in that category; welcome to the 21st century :( Fortunately there is a painless generic solution...

In my email client/reader I simply press the button marked "Junk". Thereafter the bayeisan filter follows my rule and moves the email to my "Junk folder". Every few days I scan the folder to check there's no false positives, and delete everything in that folder.

I get maybe 1 or 2 false positives a year. I order the emails in the Junk folder by sender, so oddities (which might be false positives) stand out.

They send me the paper copy, in a plastic bag. Being a good citizen, I have to remove the bag before I toss the mag into the recycling bin.

It's depressing to see what nonsense they waste our money on.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Not if it's silicon or InGaAs and has reverse bias, it won't. There are slow mechanisms that cause the quantum efficiency to increase by about

1% at higher photocurrents, but they shouldn't affect a 1-MHz signal.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

On a sunny day (Fri, 12 May 2017 07:59:40 -0700) it happened John Larkin wrote in :

I use different email addresses for different 'entities' so I see who is the spammer (what shop, etc), or who sold my email address, also I can set an auto-reply for that email with for example a request to fix their stuff, or just redirect it to trash automatically.

1+

People would be surprised at the companies who share your email addresses.

--

Rick C

The PD is biased at 5 volts, for no good reason. It's at the end of a

100 ohm shielded twisted-pair. Peak PD current can be as high as 50 mA. The voltage across the PD flails wildly when the light pulse hits, and that varicap-modulates the roughly 300 pF PD capacitance. I Spiced all that.

We should have brought in a photodiode guru consultant. I tried.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

But they send the print copy.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

On a sunny day (Fri, 12 May 2017 09:44:29 -0700) it happened John Larkin wrote in :

Print a few stickers 'return to sender, address unknown, no such number, no such phone' (from Elvis), once you get near a post box, throw it in.

Now THEY have a problem.

Sounds like w/o the bias the PD might have become forward biased during a fast pulse of light. (but I'm no guru.)

George h.

Photodiodes on cables are bad juju, for sure.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Bootstrapping would reduce the varicap distortion, but it's not appealing at the end of a cable in a system spec'd for 100 MHz bandwidth.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Jacking it up to 30V or so would be the other approach. The QL01 uses

32V. (I may have a real live licensing customer for it--big biomed outfit.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

The feedback can theoretically get you to -120 db. BUT the dip is so damn narrow that you may find temp variations on the parts moving that dip more than you may like. I do not know how one could make it from a narrow dip into a flatband response; that would be ideal.

Crude filter..phase retard with feedback?

I second this; the Q depends partly on wire resistance, and the ferrite makes for shorter and fatter wires. At 1.0 MHz, it's easy to get good inductors off the shelf. Don't know about the right tuning capacitors, though (Coilcraft 144-10J12L with ferrite tuning slug wants about 10 uF to tune to 1 MHz).

I suppose I could sum several ARB channels, to do active harmonic nulling on the generator side, the DUT input. Maybe 100% of the main signal summed with 0.2% of the distortion correctors. I think some RF systems, cell phone towers maybe, pre-distort the input to their final amps, to correct for distortion. The math would be interesting.

The DUT output still needs help, notch or highpass, before I can analyze it. I'll use one of our DDS-based ARBs as the starting sine wave, so I can cheat and tweak the frequency slightly to find the null of a notch filter. So all I need to do is tweak the depth, which it looks like one trimpot will do on a twin-tee.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

I see that 144-10J12L as ~ 450nH, it has a Q of 162 (min). Not knowing the AC resistance I'll us the DC resistance of 0.0185 ohms and I'm sure it would be higher.

It would take 0.056uf to resonate 450nH at 1MHz. The reactance of 450hH at 1MHz is 2.8 ohms I measured a 0.05uf cap as 3.23 ohms of ESR @ 100kHz.

3.23 ohms plus 0.0185 ohms = 3.249 ohms 2.8 / 3.249 = 0.862 So the circuit Q is 0.862.

Please check my work.

I don't know if this would have any relevance, but it's a filter design that knocks the second harmonic down an extra 25db over a 7 element Chebychev, uses 1 extra part.

Mikek

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus

Don't do that. The NSA might be listening and they'll leak it to DHS and the IRS. You'll never get it all straightened out.