One of my interns connected a 3KV power supply to the bias input of my Boonton 72 c-meter. Blew it up of course. The summer was almost over so I didn't bother to fire him.
Some other interns were afraid to touch 5 volts.
Show them spot welding at 1.5 volt (perhaps a forklift battery is handy?) and ask them "What me worry?" about HV...
John ;-#)#
People who are afraid of everything make bad engineers. Internship is a good way to learn about people.
Clever idea! I like it.
Yeah, I was needing this last summer for a particular problem. We have 4 PNA's but three are tied up in the microcircuit lab, and the other PNA was in for repair. The PNAs are limited to 500 mA --- that is fine for most MMIC bias.
VNAs available to me at that time were an E5071 with no built in tees and an E5061 but with no back panel access. The E5061has a voltage bias, not current, which at the time I thought wouldn't work for me because of the fixture I had and the bead approaches 0 ohms R. Now I am thinking I can get what I need with the E5061regardless of that, if I make a couple reasonable assumptions, and change my fixture accordingly.
Without some special effort, the VNAs aren't too accurate as the impedance diverges far from 50 ohms, but practically speaking it isn't so important for my application--the Z just needs to be "high" relative to 50 ohms. CM has a nice article:
The feed is probably conical. I don't know really. Conicals aren't as bad as beads re saturation, but the performance of the tee with "high" current should be discoverable regardless.
Lots of MMICs these days are self-biasing, meaning that you power them with a stiff voltage supply and an inductor. They self-regulate their current. These types have terrible low-frequency response because they have an internal dc feedback network with poorly defined lowpass elements.
I use classic klunky darlington types for pulse work.
On Fri, 07 May 2021 15:01:54 -0700, John Larkin wrote:
Something like this?
Version 4 SHEET 1 1236 680 WIRE 80 48 -64 48 WIRE 192 48 80 48 WIRE 192 112 192 48 WIRE -64 160 -64 48 WIRE 80 160 80 48 WIRE 192 192 192 176 WIRE 448 192 192 192 WIRE 736 192 448 192 WIRE 192 208 192 192 WIRE 448 224 448 192 WIRE 736 224 736 192 WIRE -64 304 -64 240 WIRE 80 304 80 224 WIRE 192 304 192 272 WIRE 736 368 736 304 WIRE 768 368 736 368 WIRE 448 384 448 304 WIRE 736 384 736 368 WIRE 448 480 448 464 WIRE 736 480 736 464 FLAG 192 304 0 FLAG 80 304 0 FLAG -64 304 0 FLAG 448 480 0 FLAG 736 480 0 FLAG 768 368 MEAS SYMBOL FerriteBead 192 144 R0 SYMATTR InstName L1 SYMATTR Value 200n SYMATTR SpiceLine Ipk=0.55 Rser=0.2 Rpar=408 Cpar=585f mfg="Würth Elektronik" pn="742792621 WE-CBF 0603" SYMBOL cap 64 160 R0 SYMATTR InstName C1 SYMATTR Value 100µ SYMATTR SpiceLine Rser=1u SYMBOL current -64 240 R180 WINDOW 0 24 80 Left 2 WINDOW 3 24 0 Left 2 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName I1 SYMATTR Value {I_test} SYMBOL voltage 448 368 R0 WINDOW 123 0 0 Left 0 WINDOW 39 24 124 Left 2 SYMATTR InstName V1 SYMATTR Value AC 1 SYMATTR SpiceLine Rser=50 SYMBOL res 720 368 R0 SYMATTR InstName R1 SYMATTR Value 50 SYMBOL res 432 208 R0 SYMATTR InstName R2 SYMATTR Value 453 SYMBOL res 720 208 R0 SYMATTR InstName R3 SYMATTR Value 453 SYMBOL FerriteBead 192 240 R0 SYMATTR InstName L2 SYMATTR Value 200n SYMATTR SpiceLine Ipk=0.55 Rser=0.2 Rpar=408 Cpar=585f mfg="Würth Elektronik" pn="742792621 WE-CBF 0603" TEXT -360 376 Left 2 !.step param I_test list 100m 500m TEXT -362 336 Left 2 !.ac dec 100 1Meg 1G TEXT 624 528 Left 2 ;SPECTRUM ANALYSER TEXT 320 304 Left 2 ;TG OUT TEXT 808 304 Left 2 ;RF IN TEXT 784 56 Left 2 ;FERRITE BEAD SATURATION TEST JIG\nRB 2021-05-10 RECTANGLE Normal 880 544 368 320 2 RECTANGLE Normal 384 336 576 528 2
The sim gives an output that sort-of matches the data sheet impedance curve. It doesn't change with test current, which is to be expected as the bead model doesn't include saturation.
C1 would be an array of low-inductance ceramics in parallel if I were to build this on copper-clad.
I'll give it a try when I get a minute. A lot of designs I have seen run the bead near its max DC rating, and I suspect a zero ohm link would be about as useful. I'll post some results if they show anything interesting.
That's the idea. I do it with power inductors too; the inductor is its own bias tee.
To measure impedance, I usually connect a 50 ohm function generator and a scope and twiddle the frequency to find the 3dB point, and maybe resonances. I TDR tiny inductors.
A single inductor connected to a scope and a function generator can show distortion when the AC current gets up, or use a function generator with DC offset to test a single inductor at modest currents. My B&K can output 200 mA.
A fungen with offset can quickly measure a ceramic cap c-v curve or diode dynamic impedance.
A lot of people were asking for "active bias" integrated into the MMIC. Certain things were thus simplified and power consumption possibly reduced. Don't blame me.
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.