4
\$\begingroup\$

I'm using this multimeter to examine a heat gun fan. The fan has four contacts placed in a circle (well in a square inside a circle). Two opposite each other are labeled + and -; the other two are unlabeled. When I extracted the fan from the heat gun, the AC inputs were connected to the unlabeled contacts.

Visible on the other side of this circular piece of PCB are four black cylinders with silver stripes at one end. Each end of a cylinder connects to a contact from a different pair (i.e. they go around the circumference of the squared circle, forming the "edges" of the square).

My primitive understanding of electronics leads me to conclude that these cylinders are diodes arranged so as to create a full-wave rectifier, which I suppose makes sense since they are regulating the powering of a fan motor.

When I measure across these contact points using the multimeter in diode mode, I get values of 500-700 for adjacent (non-paired) contacts and 1200-1400 for the AC contacts (consistently varying by about 100 depending on polarity). Conversely, measuring the resistance across the DC contacts gives about 6 ohms.

Are these diode-mode readings normal? What conclusions can be drawn from them? Is it possible to determine from them whether or not there is a continuity problem in the circuitry that would prevent the fan from running on AC power? I have tested it with success by hooking a 9V battery up to both pairs of contacts.

\$\endgroup\$
6
\$\begingroup\$

Your description sounds exactly like a full wave bridge rectifier.

A multimeter in diode mode usually displays the voltage drop measured across the diode under test, as a small current is passed through it.

A single silicon diode usually has a forward bias voltage drop around 0.7V, so your reading of 500 to 700 probably corresponds to 500 to 700 milivolts.

When you measure across the AC input, you'll be measuring 2 diodes in series, plus the series resistance of the motor winding. Your measurement of 1200 to 1400 mV still makes perfect sense.

\$\endgroup\$
  • 2
    \$\begingroup\$ @anindo Thanks for the edit. Apparently I had something other than diodes on my mind when I wrote that. :) \$\endgroup\$ – jwygralak67 Sep 3 '13 at 20:41
  • \$\begingroup\$ Cool! I'm still confused about why diode mode output sometimes displays values identical or similar to those displayed by the 200-ohm resistance mode. This seems to happen when it's being used on a circuit that does not contain a diode, or if there is a parallel circuit around the diode. It's also confusing that the mode is located in the resistance section of the mode selection dial. Is my multimeter unusual in this regard? \$\endgroup\$ – intuited Sep 4 '13 at 2:24
  • \$\begingroup\$ @intuited - both the resistance and diode test measurements involve the meter applying power to the circuit and measuring the current/voltage relationship through the element under test; the difference is more in the optimization of the measurement and how the results are presented. And on a meter without a diode test function, you use the resistance function and must interpret it yourslef from knowledge of the meter - though if you only need to identify the polarity / determine if you have a diode, you do that by comparing the readings while swapping the leads. \$\endgroup\$ – Chris Stratton Sep 4 '13 at 2:59
  • \$\begingroup\$ @ChrisStratton: I guess I'm asking why the results are presented differently. After reading allaboutcircuits.com/vol_3/chpt_3/2.html this makes more sense. \$\endgroup\$ – intuited Sep 4 '13 at 5:41
  • \$\begingroup\$ The diode test results are intended to be more directly useful by giving the voltage drop at some small test current; to numerically interpret ohmmeter results for a diode you must know additional implementation details of the meter, since a diode is not the sort of linear resistor an ohm meter is designed to measure. Incidentally, on a lot of analog meters, the lead polarity in ohm meter mode is reversed, too. \$\endgroup\$ – Chris Stratton Sep 4 '13 at 5:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.