what is the inductance of a typical 1/4 or 1/2 watt resistor? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

what is the inductance of a typical 1/4 or 1/2 watt resistor?

1 2 next

Move Thread

LAN_403

Patrick

Thu Jan 14 2010, 07:27AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Hey, everybody!

i am building a 4 transistor h-bridge to operate at 50-120khz to drive a flyback type transformer. i will use a ixys IXBD4410 and 4411 which are 14 pin dips, no gate transfomers, then 4 to-247 type igbts. My question is what about the damping resistor for the gate ive seen 1-10 ohms i want the lowest value possible to damp the ringing, the trace from driver to gate is on a double sided circuit board is 1.7 inches long and measures less then 100nH but when i add a 1 ohm 1/4 carbon composistion the inductance jumps to 1.4 uH, So what kind of resistors do you guys use? Is 1.4uH a large value for a TO-247 type circuit? i am most worried about ringing i havent got the circuit totally etched yet, so i need to know how much room to leave for a resistor should it be needed for damping.

I see Steve Ward uses a lot of 1/2 watt carbon comps on his h-bridges, do 1/2 watt resistors have less inductance?

-Patrick

Dr. Slack

Thu Jan 14 2010, 08:06AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

1.4uH is far too much to be seeing on an "ordinary" resistor, so something else is wrong. Either your measuring setup is askew, or you don't have a carbon but a wirewound, or a very spiral cut carbon-film or metal-film. How are you measuring or infering the inductance?

Bulk carbon should be very low inductance. There will be vanishingly small differences in inductance between 1/4 watt and 1/2 watt.

We use "low inductance" metal film resistors in our circuits. The difference is that they are specified to be trimmed by longitudinal laser cuts, rather than circumferential or sprial. However using bulk carbon should be just as good, at the expense of tolerance.

radiotech

Thu Jan 14 2010, 12:46PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Some issues on old resistors with metalic pigments in the colorbands -silver caused issues at RF . Also in lower ranges 1/2 watts a wider firstband
can denote wirewound. Just a trick- connect resistor to dmm on lowest voltage range and heat one end of lead close to body with soldering iron to see if it acts like thermocouple. (leadwire-resistancewire junction). If you have extras - break one apart and see how its made.

Antonio

Thu Jan 14 2010, 01:07PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

For straight lines, no matter if there is a resistor or a capacitor as part of the line, the inductance is about 1 nH/mm, or 1 uH/m for usual dimensions. Of course, wirewound or coiled resistors add their inductances to this value.

Thu Jan 14 2010, 03:18PM

Registered Member #2372 Joined:
Location:
Posts: 62

I recently measured the inductance of some 0.2 and 0.1 mOhm 1/4 watt resistors. I measured the resistance at 3 different frequencies and then found the intercept when f=>0. Z=R+iwL, so the slope is the inductace. The value that was measured was 20nH and that was the same as the calculated value, using several different methods i.e. straight wire of some thickness, conducting tube with a very thin wall, etc. They all give a result of about 20 nH.

Patrick

Thu Jan 14 2010, 03:46PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Yes, my setup gave me an unusually high number for inductance i am suspicous of the resistence (1 ohm) interfereing with my BK Precision 875B LCR meter, its usually pretty acuurate, but some circuits do test way to high or way to low. presumabaly the meter's signal is being screwed with by unwanted factors. So i measure component like resistors by themselves.

The fact that several of you say the 1.4uH is way to high, makes me think my meter is seeing to much of something else not the real inductance.

And yes i would use metal film resistors if i had any. but carbon comps seem ok to many builders, and are plentyful. ok, also, that 20-50nH sounds way more believable, and i am familliar with the 1nH per mm rule/estimate.

I feel better. I will commence the CNC cutting of the board.

-Patrick

Antonio

Thu Jan 14 2010, 08:11PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

The meter may be measuring the inductance of the test leads, if it doesn't have a compensation for their effect.

Mattski

Thu Jan 14 2010, 11:38PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

The rule of thumb I was taught in my microwave class was that a leaded resistor (solid or film composition, not wirewound or spiral cut) has around a 10nH series inductance.

Using SMD components will typically give you the best performance since you eliminate the series inductance of the resistor leads, though at 100kHz it shouldn't matter too much, especially if you keep the leads short.

Patrick

Fri Jan 15 2010, 05:37AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

It does have compensation, a good zero knob, and reads from 200.0uH to 0.0 uH (Those zeros are significant in the tenths place, so dont drop them) it has math to describe its internal series or parallel measurment and math correction for higher accuracy with the dissipation factor feature to 1khz and the 10khz and such .I like the 875B, it really makes my life easier, no more guessing or confusion on a coil i just desoldered or rewound. though i do have to be careful with the wave shape and drive conditions as inductance can be mis-measured easilly.

and Mattski what university are you at?

-Patrick

Mattski

Fri Jan 15 2010, 07:03AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

It's probably below the meter's range then. If you short the test leads does it show an an out of range error, or does it at least display 0? What about if you put the test leads across about an inch long piece of wire? I would think a bench LCR meter could compensate for resistance, but I don't really know.

It's a little tricky to test it in other ways, one simple thing you can do is put the 1ohm resistor in series with a larger resistor, drive it with a sine wave at a few frequencies, and use the voltage divider equation with Zl = 2*pi*f*L_series. But you have a problem in that the other resistor will also have a similar inductance, presumably, and so you make it larger (say 20ohms) to reduce the error, but now the voltage you measure is harder to measure accurately. There are bridge methods, but those are kind of complicated and depend on having accurate parts in the bridge. The easiest way of course is to measure on a network analyzer, but it's kind of overkill :)

I go to UC Santa Barbara by the way. Are you a student currently?

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob