Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. The sunn still shines online!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  
previous: Dave Stephens Moocow: Wow am I glad you are back!... -- 8/26/2004 11:26 AM view thread

Re: Moocow's Inductance Test, flawed math

8/27/2004 4:30 PM
Joe Gwinn
Re: Moocow's Inductance Test, flawed math
I have not tested widely, but many of the inexpensive LCR meters only work well with relatively "pure" components. In the case of inductors, if the "Q" is low (as with pickups), the reading will be wildly off. The higher the DC resistance of an inductor, the lower the Q. Numerically, Q is the reactance divided by the resistance, at the test frequency f.  
Q= (2 Pi f)/(resistance)  
Q stands for Quality, as in purity of the component: a high-Q inductor will have low resistance and capacitance, and so on.  
Specific example: I have a BK Precision model 875B LCR meter that I got on sale for about $200. Measuring the primary of a small audio transistor output transformer, the 875B reports 1.08 Henrys, which is quite plausible. The DC resistance is 17.5 ohms. Then I added a 2500-ohm pot in series with the primary. As I increased the series resistance, the indicated inductance grew, slowly at first, then explosively, indicating 66.5 Henrys while 2,716 ohms of added serial resistance was present.  
The 875B uses a 120-Hz signal to test inductances exceeding 200 millihenrys, and 1,000 Hz for 200 mH and less, so f=120 Hz here.  
The reactance of a 1.08-H inductor at 120 Hz is 2*Pi*120= 754 ohms, and the Q without the added resistance is 754/17.5= 43, which is reasonable as transformers go. When the added resistance dropped the Q to about 5, the measurement error was about 30%. As the Q went below 5, the error exploded.  
So, for a pickup with say 4 H inductance and 8 K series resistance, the Q would be 3016/8000= 0.38, where the 875B is totally useless.  
I suspect that the Wavetek Meterman suffers from the same disease; this is easily tested.

Dave Stephens Greg: I think connecting my Wavetek... -- 8/28/2004 2:10 AM