NAD Idling Current Adjustment - Help? What? NAD 3225PE

[SanJoseMichael]

Sorry to be such a noob, especially at my undisclosed age. Came across an AK posting (Amplifier Distortion, DC-Offset, and You!) which led me to dig out the Fluke meter and go wandering inside an old Denon and a pair of NADs I've been running lately. What follows is from the service manual for the NAD 3225PE/3020I. I understand the Center Voltage part, but not the Idling Current part. See photos as well.

IMPORTANT
Speaker impedance switch should be in 8 ohm position while adjust center voltage and idling current.INITIAL ADJUSTMENT (No load connected)
INITIAL ADJUSTMENT (No load connected)
A. CENTER VOLTAGE
1. Connector DC millivoltmeter to L channel output terminal.
2. Turn on and adjust to 0 V +/-30mV with R411 (470 ohm). Connect DC millivoltmeter
to R channel output terminal and adjust R412 to 0 V +/-30mV.
B. IDLING CURRENT
1. Remove solder short across R455 and R456.
2. Connect DC millivoltmeter across R455 (1 ohm) (output transistor's collector resistor) and
adjust R443 for 26-30mV reading on meter. Repeat adjustment with R444, connecting meter
across R456 (1 ohm).
3. Leave power on for minimum 5 minutes.
FINAL ADJUSTMENT
C. CENTER VOLTAGE
1. Repeat step A about.
D. IDLING CURRENT
1. Repeat step B and adjust with R443, R444 for 30mV reading on meter.
2. After the alignment is finished, 1 ohm resistor R455, R456 is shorted by solder short.

My brain seizes up at B. IDLING CURRENT:

I see that R455 is a 1 ohm resistor, with perhaps a (missing) solder short between P402 and P403 (assuming P402 and P403 are in parallel with R455)?

Likewise R456 is a 1 ohm resistor, with perhaps a (missing) solder short between P405 and P406 (assuming P405 and P406 are in parallel with R456)?

Here, I assume I set my meter to measure millivolts, with probes on both ends of the resistor (in parallel with the resistor), adjusting with R443 and R444 respectively. Is that correct?

And, I note the solder short, however it may have been implemented, has not been replaced. What effect might that have on current performance?

Can I just tack solder a jumper over R455 and R456 on top of the board, or try to make small wire U-shapes to plug between R402 and P403, also P405 and P406?


Appreciate any guidance that can be given.

 

[NAD80]

After doing the alignment have to put back the solder short. The 1 ohm resistor is only there for the alignment only. The 1 ohm resistor could burn up if left without the short.

 

[Powertech]

For SanJoseMichael. A brief description of what idling current is. For others who know, this is a brief description for bog standard class A/B power amps. Yes I do know about class A, current dumpers and all manner of other variants.

If you can visualise the sound signal oscillating above and below a zero point (positive then negative then positive etc) many times per second. A 1kHz signal will pass through the zero point 2000 times per second. 1000 times in each direction. The output stage in a class B amplifier has 2 power transistors, one will amplify the positive part of the waveform and one will amplify the negative part of the waveform. The problem with a B class amplifier is that these transistors will only switch on when the signal is about 0.7 volts above or below zero. This means that there is a small portion of the signal around the centre of the waveform where there would be no output. As the waveform passes these points (either positively or negatively) each transistor in turn would suddenly start conducting. This would produce a nasty little step in the output signal. It would be very audible and is called crossover distortion. In order to get over this, the output transistors are given a bias signal which prevents them from switching off completely. This prevents that nasty little step. It also means that at very low signal levels the amplifier is operating in class A (I'm not going to explain why), which is why these amplifiers are called class AB. Class A with a low signal and class B with a higher signal.
Adjustment of this bias signal dictates the minimum or idle current passing through each of the output transistors to just above the point at which they would turn off completely. If this current is too low, the transistors will switch off and then switch on again each time the signal waveform crosses the zero point. If the idle current is set too high, the output transistors will overheat because the minimum current plus the higher signal currents will produce more heat than the heatsink can remove.
Each output transistor has a large low resistance high wattage resistor connected between it's emitter and the output line (usually 0.1 - 0.3 ohms). This resistor stabilises the transistor. In your case there is a second resistor which is placed temporarily in series with the emitter resistor for measurement purposes. When a current at a given voltage passes through a resistor, that current produces a voltage drop across the resistor. By measuring this volt drop you can calculate the current flowing through it using ohms law. The manufacturer has done the calculation for you and told you the volt drop that they require across this resistor to give the correct quiescent (idling) current required through each output transistor. Because you are setting a minimum current, this setup has to be done when there is no signal passing through the amplifier.
With most amplifiers, the idle current setup is done by measuring the volt drop across the emitter resistor. You then have to do the calculation to work out the idling current. Most manufactures err of the low side to prevent amplifier overheating and maintain reliability. The optimum for best sound quality is usually somewhat higher then this, but if you are unsure, just stick to what they say. As a guide, don't set below the value given, but is you are a little higher it isn't critical and will not damage the amplifier. I might run a little bit warmer though.

I hope this helps.

 

[Powertech]

Just as a follow up.
Out of interest for those that wish to fiddle. Be careful - there are some nasty voltages inside an amplifier. If you are unsure DON'T mess with it!.

The quiescent current is obtained by dividing the volt drop across the emitter resistor with the resistance value.
For example, if the emitter resistor is 0.22 ohms and you have 10mV dropping across it - 0.01volts(10mV) divided by 0.22ohms = 45.45 (mA).

My Rotel RB976 6 channel power amp runs with a quiescent current of 90mA because it sounds better. This is way above the manufacturers' recommended value. It also runs a bit hotter than standard.
BUT - and this is a big but:
1) It has thermal protection from a thermistor mounted on each heatsink between the output transistors.
2) The thermal shutdown is set to activate if a heatsink temperature exceeds 80degC.
3) The amp has a full blown electronic monitoring system which will shutdown the both the unit power supplies and disconnect the speakers on excess temperature, excess emitter current and high DC voltage detection at the output.
4) The power transistors have been replaced with 25amp output devices as opposed to the original 15amp transistors (most 60 watt amps only use 12amp transistors anyway). The temperature protection is set to lie well within the transistor safe operating area (SOAR).
5) It's kept well ventilated so that it can remove excess heat.

 

[SanJoseMichael]

Powertech, thanks for an excellent description, it sounds like push-pull in my old Magnavox tube amps.

Okay, I will measure and adjust across the resistors.

But, where were the resistors originally shorted (if they were at all)? Would I have been across the resistor itself or though R402 and P403, also P405 and P406? I looks like I'd have to pull the board in order to get to the solder side underneath in order to install a jumper across R402 to P403, P405 to P406, so solder tacking a wire across the resistors would be much, much easier.

 

[TechWazz]

Hello PowerTech

I have a Rotel RB-976 Amplifier as well.
what is the Emitter Resistor Value of the Rotel?
The biasing pots allows for us to adjust in millivolts, am I right to assume that we adjust the pots accordingly to the voltage required to allow us to have our desired quiescent current, which is calculated by the readings of the bias voltage divided by the emitter resistor value?

Eg: Measurement of TP1 and TP3 which is in millivolts, bias trimmed by VR601.

I will be very grateful if you can advise me on this.

 

[cabby]

"Remove solder short "
Does this mean removing the solder from the component and taking out the resistor from the board before measurement? Then once tweaking is completed, resolder using soldering iron ?

 

[petehall347]

remove solder blob on backside of board .

 

[cabby]

remove solder blob on backside of board .

ok, so basically removing it from the circuit, correct?

 

[petehall347]

yes .doing so brings the emitter resistors into the circuit .

 

[mbz]

I would expect a solder blob/bridge at the following locations. Just remove with solder pump or whatever.
As previously mentioned, it must be put back once the idle has been set.

 

[Goldie99]

ok, so basically removing it from the circuit, correct?

Assuming we're still talking about the NAD 3225PE, and that the published schematics are correct - no. The R455 & R456 resistors should not be removed from the pcb at all.

The NAD 3255PE seems to be similar to the 3020 models, in that it doesn't use emitter resistors. The R455 & R456 resistors are actually 1 Ohm resistors which are permanently inserted into the collector feed to the output transistors, for the sole purpose of setting the idle currents.

In normally operation, the solder "shorts" are simply blobs of solder which connect both sides of the R455 resistor together (R456 is analogous), on the solder side of the pcb normally - in other words, R455 & R456 serve no purpose in normal operation, they are shorted out by the solder blobs, and effectively form no part of the circuit (but are still there for next time they might be needed).

When the idle currents are to be adjusted, then the solder blobs are removed, and R455 / R456 become part of the circuits - they are simply 1 Ohm resistors, in series with the (+ve) collector feed to the outputs - when you set the idle currents, and adjust for 26 - 30 mV across R455 or R456, you are simply setting a 26 - 30 mA current in the (+ve) collector feed to the outputs.

Once the idle currents have been set - don't forget to replace the solder shorts !

 

[northpaw]

Here are what the solder shorts actually look like, on the underside of a 7220PE PCB (very similar to the 3225PE, but with a slightly different numbering scheme), and the schematic for a section of the left-side amp with the shorts shown by dotted lines. All relevant areas are in the red rectangles.

Even though not stated in the manual, it seems you can use the pairs of test points (P40x) to measure across instead of clipping onto the resistor leads themselves. That is, for the 3225PE numbering scheme, across P402-P403 for R455, and across P404-P405 for R456. Since these are not marked on the schematic, before doing so, verify this by testing for 0 ohms across those points with the shorts still in place, or 0 ohms between each test point and the appropriate resistor lead (you will need a decent meter to tell this is actually different than 1 ohm).

When I did this on the 7220PE, I removed the solder shorts and then replaced them after setting the bias, but it seemed to me at the time, that instead of replacing the solder shorts, I could have placed substantial slip-on jumpers across each pair of test points (i.e., P402-P403 and P404-P405 in your case), but decided not to.

I agree that this set up with solder shorts is cumbersome, and a deterrent to setting the bias on these NAD units.

 

[cabby]

Assuming we're still talking about the NAD 3225PE, and that the published schematics are correct - no. The R455 & R456 resistors should not be removed from the pcb at all.

The NAD 3255PE seems to be similar to the 3020 models, in that it doesn't use emitter resistors. The R455 & R456 resistors are actually 1 Ohm resistors which are permanently inserted into the collector feed to the output transistors, for the sole purpose of setting the idle currents.

In normally operation, the solder "shorts" are simply blobs of solder which connect both sides of the R455 resistor together (R456 is analogous), on the solder side of the pcb normally - in other words, R455 & R456 serve no purpose in normal operation, they are shorted out by the solder blobs, and effectively form no part of the circuit (but are still there for next time they might be needed).

When the idle currents are to be adjusted, then the solder blobs are removed, and R455 / R456 become part of the circuits - they are simply 1 Ohm resistors, in series with the (+ve) collector feed to the outputs - when you set the idle currents, and adjust for 26 - 30 mV across R455 or R456, you are simply setting a 26 - 30 mA current in the (+ve) collector feed to the outputs.

Once the idle currents have been set - don't forget to replace the solder shorts !

Well, mine is the 7240PE , but noticed the service manual instructions were very similar to this item for the OP.

 

[cabby]

I have the 7240pe but will piggyback this thread secondary to having same problem. The photo shows the blobs that I unsoldered which is how I understood " removing solder short " per the service manual. Dvm across the resistor leads in blue reads 0mV and the corresponding VR does nothing when turned. Ditto for the other channel. What am I doing wrong ?

 

[northpaw]

What you have labeled as "blobs" do not look to be a likely a solder short location, in my experience with other NAD 72xxPE models. The two things on the end of the black lines look like empty solder pads. And there should be just one "blob" per channel. See the photo in my earlier post on a 7220PE above, where you can see a single little mound of solder that bridges across the surface traces on the PCB, one for each channel (highlighted in the red boxes). Things may not be just like that on your unit, but I would have thought they would look somewhat similar.

Since the SM for the 7240PE does not show shading of the traces on the layout diagram, I can't verify myself that the location you have is correct. You can do that by carefully inspecting both sides of the board to make sure you have it right. Shining a bright flashlight on the component side while looking at the solder side of the PCB can help in locating yourself more confidently (you will see the shadow of the components). It often takes multiple attempts to be sure you have it right. Once you do, you can further verify by measuring the Ω between the test points. You should get 1Ω (however, many inexpensive meters may have difficulty getting a reading with this low an resistance).

Before you proceed, if you have not already done it, you should be careful to return the two VRs to their original settings. This will prevent starting up with too much bias and harming your transistors.

 

[mbz]

The solder blobs short out resistors R471, R472, these are 1 ohm resistors.
See pick below to locate R472.

 

[mbz]

 

[cabby]

The solder blobs short out resistors R471, R472, these are 1 ohm resistors.
See pick below to locate R472.

Hi, yes I've located these and stared at them quite a bit last night....R472 in particular is hard to get to, crammed right next to the transistor.

What you have labeled as "blobs" do not look to be a likely a solder short location, in my experience with other NAD 72xxPE models. The two things on the end of the black lines look like empty solder pads. And there should be just one "blob" per channel. See the photo in my earlier post on a 7220PE above, where you can see a single little mound of solder that bridges across the surface traces on the PCB, one for each channel (highlighted in the red boxes). Things may not be just like that on your unit, but I would have thought they would look somewhat similar.

Since the SM for the 7240PE does not show shading of the traces on the layout diagram, I can't verify myself that the location you have is correct. You can do that by carefully inspecting both sides of the board to make sure you have it right. Shining a bright flashlight on the component side while looking at the solder side of the PCB can help in locating yourself more confidently (you will see the shadow of the components). It often takes multiple attempts to be sure you have it right. Once you do, you can further verify by measuring the Ω between the test points. You should get 1Ω (however, many inexpensive meters may have difficulty getting a reading with this low an resistance).

Before you proceed, if you have not already done it, you should be careful to return the two VRs to their original settings. This will prevent starting up with too much bias and harming your transistors.

Thanks, I did turn just a bit here and there, and did not crank anything. None the less, doesn't feel good not knowing what the heck the idle is at now. I'll have to check again, but I'm pretty sure I've located the right solder on these.

 

[cabby]

Here is the other resistor.