bias setting

[ben goh]

Hi

Need help and clearance.

In idle mode with 4 ohms load , the bias setting on the emitter

resistor 0.5 ohms is 10mv. If a 1 khz of 200 mv is fed, the bias

went up to 40mv with a constant output signal of 5.6 volts a.c.

Is the bias normal?

Thank you for any feedback.

[Netlist]

Ben,
This looks fine to me. It is normal that the current increases when signal is applied. Of course, without knowing the amp, it's impossible to tell if the values are within specs.

/Hugo

[AndrewT]

Hi Ben,
it is normal to measure the bias with the speaker disconnected.

Otherwise any DC output offset will send some current into your load.
Then the bias currents in the opposite resistors will be different. So neither reading (voltage) gives a true bias current.

When your multimeter is reading 40mV is it the same for both emitter resistors?

Can your multimeter differentiate between DC bias voltage and AC output current (and voltage) flowing through the emitter resistor?

[ben goh]

Hi

I have two digital meter to test. One on the bias and the other

on the output. Both on d.c volts range. 0 volts dc is achieved

at the output and when I cut the 1khz tone, the bias went back

to 10mv. It is the same with the others. My power output are all

NPN transistors. Please enlightened me. how to measure the

ac output current.

Thanks

[anatech]

Hi Ben,
Bias current is not normally measured dynamically. Bias current is always measured with no signal and no load (open).

Your current reading will increase with signal as Hugo said. How much is not known. It will depend on your meter and the amp. The impedance (not resistance) of your load at whatever frequency is important.

If you wish to measure the AC current in the load, you can use a current clamp for your oscilloscope (around one lead to the speaker), a sensing load in series with the load resistor or an accurate, resistive load (non-inductive). For the resistive load, just measure the AC voltage, most Digital meters are inaccurate over a couple hundred Hz. An Agilent or Fluke brand is what I could trust. Almost no other brand is worthwhile for this. An oscilloscope may not be accurate enough.

-Chris

[ben goh]

Hi Chris

Thanks for the infor.

I am using Fluke 79 meter for the test.

Without load and signal, the bias across each resistor is at

constant 10mv.

I presumed the amp should be working fine.

Thanks

[anatech]

Hi Ben,
I'd say so.

I good test is to measure the bias cold, then after about 5 min. Then heat the heasink up with tones or music into a dummy load, then disconnect and measure the bias current with the heatsink hot. I like to see a slight reduction in bias current. Most times you will see a slight positive change. That's okay too. A large positive change means that the amp may go into thermal runaway if ventillation is reduced and music is played loud. This is really pretty common in a household.

-Chris

[AndrewT]

Hi,

measure the bias current with the heatsink hot. I like to see a slight reduction in bias current

I second that. Reducing bias as the amp overheats can only help longevity.
The increase in distortion might even become noticeable and tell the listen to switch off.

[anatech]

Hi Andrew,

The increase in distortion might even become noticeable and tell the listen to switch off.

Not that much of a reduction!

When you test an amplifier, the top cover may be off. Also the amp is not on a shelf. Once th ecover is on, and the amp has less airflow, that negative change may become a positive one. An increase in bias current can only become worse once installed.

Interestingly, most Marantz amplifiers exhibit a slight negative change in bias current with increasing temperature. Looking at the circuits, you know this was intentional. I feel they made the correct choice. The change in bias is never enough to affect the distortion measurements.

If I design an amplifier, I think one test will be the installation horror of those small stereo stands with the glass doors.

-Chris