How do YOU set bias?

[z-adamson]

When you have a spec and measurement instructions, sometimes it will also say WHEN to check it.

For example, "2mvdc at test points.....take measurement 2 minutes after start up because as the amp continues to warm up the measurement will be higher".

And other times it will just say "2mvdc at test points". No mention of how much run time before measuring. In this case, do you wait till the amp is 100% up to temp? Do you measure right after firing up? 5 min after start up?

 

[Powertech]

This has been covered before, but to reiterate-
1. The amp should be at normal working temperature when idling (switched on with no signal or speakers connected).
2. There may be test points - If so, use them.
3. If there are no test points you will need to locate the output transistor emitter resistors. Either way, you are measuring the voltage drop across them.
4. Use miniature sprung hook test leads to prevent contact with other components or internal fittings - misconnection could blow the amp up!
5. Switch amp off when making test connections. Check. Check again and then switch the amp back on.
6. Measure the volt drop and allow time to stabilise - keep meter on for a few minutes.
7. Adjust the bias pot VERY SLOWLY to manufacturer's spec. Do this in small steps and allow the amp to stabilise after each step. The setting is often quite sensitive.
8. Retest the values after about 30 minutes.

Even though you are measuring voltage, what you are actually setting is the minimum emitter current which is allowed to flow through each output transistor with the amplifier at idle. As the current increases through the resistor the voltage will increase in direct proportion to this.
There is a formula to use if current is given instead of voltage in the specification:-

Emitter current = Emitter resistor volt drop (volts) divided by the emitter resistor value(ohms) e.g 0.01(10mV) / 0.22ohms =45.45 mA
The emitter current in a class AB amplifier is normally somewhere between 20mA to 50mA, but it can be higher in some circumstances.

 

[z-adamson]

This has been covered before, but to reiterate-
1. The amp should be at normal working temperature when idling (switched on with no signal or speakers connected).
2. There may be test points - If so, use them.
3. If there are no test points you will need to locate the output transistor emitter resistors. Either way, you are measuring the voltage drop across them.
4. Use miniature sprung hook test leads to prevent contact with other components or internal fittings - misconnection could blow the amp up!
5. Switch amp off when making test connections. Check. Check again and then switch the amp back on.
6. Measure the volt drop and allow time to stabilise - keep meter on for a few minutes.
7. Adjust the bias pot VERY SLOWLY to manufacturer's spec. Do this in small steps and allow the amp to stabilise after each step. The setting is often quite sensitive.
8. Retest the values after about 30 minutes.

Even though you are measuring voltage, what you are actually setting is the minimum emitter current which is allowed to flow through each output transistor with the amplifier at idle. As the current increases through the resistor the voltage will increase in direct proportion to this.
There is a formula to use if current is given instead of voltage in the specification:-

Emitter current = Emitter resistor volt drop (volts) divided by the emitter resistor value(ohms) e.g 0.01(10mV) / 0.22ohms =45.45 mA
The emitter current in a class AB amplifier is normally somewhere between 20mA to 50mA, but it can be higher in some circumstances.

I have never had an amp without some kind of test points....

But, I basically do as you say other than the VERY SLOWLY part. I never really paid that close of attention to the rate at which I turn the pot and typically I am using bourns 20 turn pots.

When I do this and I have the amp running, I will reach a point where the bias is no longer drifting around. I define this as the amp becoming stable. I will adjust to spec. I will wait for a while for it to stabilize again. Once I have it stable and to spec, I idle the amp for quite a while monitoring bias just to confirm that it does not start to drift. After that, I will play music through it for a couple days, then check bias again just confirm that it is still to spec only to find that it is no longer within spec. For example...instead of 2mvcd + or - 0.5mvdc it will be 3 or 4mvcd at the test points.

 

[Powertech]

The guide I gave was not intended to be specific, but just a generalised format for adjusting output bias current.

I have never had an amp without some kind of test points....

I have, which was why I made the point.
The reason I said to go slowly is that most amps have 270degree open presets fitted which tend to be a bit coarse when adjusted.. It is much easier if you have multiturn pots fitted.

Generally speaking a little bit of drift won't make much difference, but it's worth checking anyway.
The main criteria which I forgot to mention was to also keep an eye on heatsink temperature and that the specs have been known to have misprints, hence the additional note about how to work out the emitter current and the normal expected ranges that are usually employed in class AB amps.

 

[Scott75s1974]

When you have a spec and measurement instructions, sometimes it will also say WHEN to check it.

For example, "2mvdc at test points.....take measurement 2 minutes after start up because as the amp continues to warm up the measurement will be higher".

And other times it will just say "2mvdc at test points". No mention of how much run time before measuring. In this case, do you wait till the amp is 100% up to temp? Do you measure right after firing up? 5 min after start up?

I'm interested in this specific part of your question as well. The careful, fully warmed up method, allowing time for any bias setting changes to stabilize is without a doubt a good method to set a known, fully warmed up steady state bias current/voltage.

But service manuals often specify a setting with a much shorter warm up period. They certainly knew as well as us that it would likely continue to rise. But they specified 2 min or 5 min. So I worry that setting bias at the SM specified level, at a much longer than SM specified time will end up underbiasing the circuit.

(My Scott r75s specifies about 25 mV at 5 min. If it let it warm up, that setting will stabilize at about 40 mV. If I dial it back to 25mV after 30 minutes, the driver and output transistors stay pretty much ice cold. I've checked with an owner of this model with factory settings on bias and they measure closer to 40mV (35/50) after full warmup.)

 

[triode17]

I mostly work on tube amps, so I will turn it on and wait for 15 minutes to pass. This is based on Morgan Jones observation that the heater takes 1 minute to reach full temperature. The cathode takes 5 mins. to do the same. And the Plate takes 10 minutes (this is my own observation) so 15 minutes seems right. I also have noted that the bias current usually will drift up and exceed the set current until it settles back down after 15 min. I then measure and multiply the Plate current by the Plate voltage to determine the Plate dissipation. It gets set to approx. 70%. I prefer 75% for a warmer sound.

 

[maxhifi]

I let it warm up before I set bias, unless there's instructions to the contrary. Maybe it's a carry over from working on tube equipment.

 

[Harvestor]

When you have a spec and measurement instructions, sometimes it will also say WHEN to check it.

For example, "2mvdc at test points.....take measurement 2 minutes after start up because as the amp continues to warm up the measurement will be higher".

And other times it will just say "2mvdc at test points". No mention of how much run time before measuring. In this case, do you wait till the amp is 100% up to temp? Do you measure right after firing up? 5 min after start up?

If you search my history dating around June it will see that I had adjusted the bias on my receiver amp section and after the first initial reset the next time I turned the receiver on I noticed the bass extensions on my left set of woofers was not the same as the right Channel with particular songs it was immediately noticed so I made about five subsequent adjustments between April and June and the last one I made some sort of mistake with the test point and my probe from my Fluke meter and it shorted out the transistor and blew up several parts the circuits related to the amp..
After speaking with the tech that worked for Tech HiFi at one time in the 1970s and early 80s he stated that the millivolts will drift always a few millivolts depending on temperature I express to him that I had discovered this and would give several adjustments over the course of an hour or two as the receiver was being used because I had noticed this..
Long story short is be 100% positive of the test point connection in my instance a miniature pair of alligator clip type probes would have saved me the grief but also in hindsight being 20-20 I should not have even adjusted in the first place but I could have at least checked once just to make sure it wasn't astronomically out of spec which it was not so now a perfectly Mint Condition 100% like new receiver of mine that I covet and cherish is in need of several hundred dollars of qualified technician to repair..
So another moral of the story is don't be too overconfident and the less you mess with something the better and if it isn't broke don't fix it ....
just an FYI Food For Thought from my experience

 

[gadget73]

I usually hook the meter up and observe it from cold, especially if this is something I'm powering for the first time after repair. If the bias starts to run away or goes way high I'll back it down but I don't try to make final adjustments until it has warmed up and settled. Transistors are prone to thermal runaway so those should be monitored for a while after adjustment to make sure it doesn't creep too much.