Beomaster 2400 Type 2902 - Canada Project Bench Testing To Collect Some Reference Data

Listening to this restored Beomaster 2400 has been very enjoyable. The amplifier sounds as good as any Beomaster 1900 and 2400 receiver I have tested in the past.  But that is a little subjective isn't it?

In my ongoing effort to collect some quantitative data so I can compare my restoration results I returned to using my QuantAsylum QA401 Audio Analyzer and some fixed resistive, dummy speaker loads.
Since I last used it QuantAsylum has released a couple of software upgrades that will help me with my amplifier testing.

If you are not familiar with the QA401 Audio Analyzer, it can output stimulus signals into an audio amplifier and measure the outputs either from preamplifier outputs (which a Beomaster 1900/2400 does not provide) or output amplifier speaker outputs.

A nice feature of the QA401 analyzer is that it can do single-ended measurements or differential measurements. I have experimented with both in the past. For this test setup I went with differential measurements. 

Choosing the measurement points based on the expected output is important.  For that reason I constructed my fixed resistor, dummy speaker loads like this -

...and here are the connections to the QA401 Audio Analyzer.

I have a dummy speaker load assembly for two amplifier output channels.
I decided on 8Ω fixed loads.  If I had to it would be quite easy to desolder and resolder my configuration for 4Ω fixed loads. 

An important reason for having multiple series load resistors in the fixed load assembly is because the QA401 inputs have limits to how much voltage they can handle. I have tested amplifiers on my bench that have 20 Watt outputs all the way to amplifiers that have 250 Watt outputs. With the Beomaster 2400 I am only looking at 20W into 8Ω.

My dummy speaker load assembly is constructed like this (from the positive lead to the negative lead):
4Ω to 0.08Ω to 2Ω to 2Ω

That gives me a total 8.08Ω for the fixed resistance load across the amplifier speaker outputs.

The 0.08Ω resistor is in there for the large 100 plus watt amplifiers. I need a small sensing resistor to connect my QA401 differential measurement probes without exceeding their input level limits.

Determining what the sensing load I am going to use is important in the configuration of the test measurement system of the QA401. It has a popup context menu for setting the Input Gain level of the measurement. If I was going to measure across the full 8.08Ω loads then I would use an Input Gain value of 0.00 dB. However, as I explained above, the QA401 has input voltage limits so it is common to be measuring across one of my individual (smaller) resistor segments. When doing that you must calculate the Input Gain value to put in.

For example if I am concerned about the amount of voltage going into my QA401 measurement then I could begin by measuring the amplifier output using the small 0.08Ω sensing resistor.
In using that I calculate my Input Gain value using the formula 20*log10(0.08/8.08)) which gives me -40dB.

Also note there is the option to display the output as Vrms or Watts.  The actual load impedance value field here is for the QA401 to calculate the values it displays. This is the impedance the amplifier sees, not the individual sensing resistor value.

Why not just use the 0.08Ω sensing resistor for all amplifier measurements?
The reason to not do that and the reason for having the other measurement point options are for sensitivity of the measurement. Basically the resolution for accuracy.  According to the QA401 documentation and forum discussions you get the most accurate measurements when your test signal is not too close to the maximum input the QA401 can handle and not too small either.

For the Beomaster 2400 and its 20 Watt power output rating across an 8Ω load I am going to make my measurements across one of the 2Ω resistors in the assembly.  That means a QA401 Input Gain setting of -12dB.

Continuing on with some actual measurements...
I wanted to first check the QA401 measurement setting configuration and my earlier service manual adjustment of the Beomaster 2400 volume level (for the Medium Volume Preset switch position).
Remember that test setup asked for a 200mV, 1KHz test sine wave into the Tape inputs.
Turning the Beomaster 2400 on using the Tape source selection should result in 100mVrms at the speaker outputs at the medium volume level setting on the Beomaster.

Here is that check.

Note that I set the QA401 measurement units to volts so I could see what it measures for the speaker outputs. The voltage output levels are in the 100mVrms range. I was hoping they would be closer but this is the first time I have recorded that setting so perhaps that is typical. I will start noting that on all of my Beomaster 1900 and 2400 restorations from now on. 

The bench tests I really want to collect data on is the total harmonic distortion (THD) at 1KHz and the frequency response. For all of these measurements I will need to use the Beomaster Tape source as the point of entry for the test signal. I decided for these tests to keep my Tape test signal at 200mV.
On the signal generator of the QA401 that calculated out to be -13.8dB for the generator configuration.

That value was determined using a generator test option on the QA401.
I set the QA401 generator amplitude value while measuring the signal the QA401 was putting out using a DMM.

The bench testing generator is now configured to go along with the measurement configuration.

I began the bench testing with the THD test option on the QA401. The generator is sending a 200mV, 1KHz test signal into the Beomaster 2400 Tape inputs. I increased the volume level on the Beomaster 2400 while watching the QA401 left and right measurements coming off the speaker load assembly.
I set the measurement units to watts for these tests and here are my results with the Beomaster producing 20 Watts of power.

Less than 0.02% THD.  That is a very good number and that is at the maximum output level for this amplifier.

I left the Beomaster 2400 at the volume level that produced the 20W output (at 1KHz) and removed the test signal.

Next, I setup the QA401 for a frequency response test using that same volume level.
I let the QA401 do a single shot sweep frequency response test from 20Hz to 20KHz and saved the result. 

Another nice measurement. The swing of the signal from 20Hz to 20KHz for each channel stays within 1.5dB from the 1KHz point. 

That is all I will collect for now. I believe I have a repeatable test configuration I can use on future Beomaster 1900 and 2400 receivers.

Beomaster 2400 Type 2902 - Canada Project Tying Up Some Loose Ends and Completing the Reassembly

The Beomaster 2400 has been playing music without a hitch since I started the listening/play tests. 

Now it is time to tie up any loose ends and complete the reassembly of the Beomaster in its cabinet.

There is one more service manual check I needed to make.  That is the audio output level setting the service manual recommends for the volume control.  The procedure uses the Tape audio source input as the reference for setting this volume level. 
The adjustment procedure calls for inputting a 200mV, 1KHz sine wave signal to the Tape source going into the Beomaster.
The volume out measurement is made with a DMM connected to the speaker outputs.
The Beomaster 2400 volume preset level set to the Medium position.
The DMM should read 100mV across either of the channel speaker outputs.  Trimmer 2R136 is used to make that adjustment. What the 2R136 trimmer is doing is adjusting the intensity of the lamp inside the Beomaster 2400 volume control module which directly controls the volume level. 

Now to finish the reassembly of this Beomaster 2400...
Here are the Bass, Treble and Balance indicator masks connected back up to the slider controls.

I also installed new plastic feet that I get from Martin Olsen at Beoparts.

This Beomaster receiver arrived with a missing Speaker 2 selection switch lever.  Luckily I had a spare and I got it installed to the bottom plate of the cabinet.

Finally all of the pieces fit back together.

Beomaster 2400 Type 2902 - Canada Project Assembly and Last Adjustments for the First Play Event

The no-load (idle current) adjustments and the power supply adjustments were made earlier.  The lamps were replaced and I performed a little test measuring the source switching volume mute circuit. Now it is time to finish up a couple of adjustments for the FM tuner, partially reassemble the Beomaster 2400 and see how it sounds.

The service manual has a couple of easy FM tuner adjustments that can be made.
One is for the FM tuning voltage on the FM tuner potentiometers. 
FM1 preset is turned to the dial stop for 88MHz and a test point (4TP1) is measured across resistor 4R1. The voltage should be set to 4.7 volts using trimmer 4R2.
The same measurement point is used again to measure FM5's potentiometer set to 88MHz. On this adjustment, trimmer 4R8 is adjusted to get 4.8 volts (across 4R1).

Another quick FM tuner adjustment is for the tuning indicator lamps. Test point 2TP3 is shorted to ground and the Beomaster 2400 is turned on by selecting an FM source.  Trimmer 2R31 is adjusted so the brightness level of the two tuning indicator lamps are equal brightness.

Finally it is time to hear what this Beomaster sounds like after all of the restoration work.
I connected up an iPod Nano to the Beomaster 2400 Tape source input, my FM antenna to the PAL FM plug and the Speaker 1 outputs to the Beovox S-55's that I keep next to the test workbench.

Standby lamp on...

Tape source on

FM1 on

The Beomaster 2400 sounds good.  The control sliders for Bass, Treble and Balance all work nice and cleanly. A nice milestone in a project to hear music coming from the restored components.

I will do more reassembly of the Beomaster 2400 cabinet and continue testing. I will test out the remote control functionality next. That includes testing out a nice Beogram 4004 as the Phono source. In case you didn't know, the Beogram 4004 can be remotely controlled from a Beomaster 2400. 

Beomaster 2400 Type 2902 - Canada Project Indicator Lamps Replaced

 The indicator lamps are now all replaced on this Beomaster 2400.  The old lamps (except for one) were still working but the age of the lamps is unknown so with this restoration it will be documented when they were changed.

The only lamps I converted to LEDs are the three lamps for the Bass, Treble and Balance position indicators.  Those lamps have no function in the circuit except to illuminate the slider positions when the Beomaster is on.  From past Beomaster 1900/2400 projects I like using two LED devices per lamp position here to give the light coverage on the indicator masks.  The reason to change these three incandescent lamps is because the give off the most heat which I have seen cause damage to the indicator masks. You can also see in the photos that the center of the red indicators lenses on these three lamps is faded white due to the heat. The replacement LEDs will run much, much cooler.

The rest of the Beomaster 2400 (and 1900) lamps do directly affect the circuits they are indicators for.
So I will replace those using replacement lamps the Beomaster owner sourced from Martin Olsen.

Here are the Volume Control and Source Select lamp boards with their new lamps.

Now with the lamps all replaced I rechecked the volume muting circuit when a source is selected.

The resulting voltage levels and volume muting duration are pretty much the same as with the old lamps. This is something I can note and use when comparing the next Beomaster 2400 (or 1900) receiver.

I am now ready to recheck the service manual adjustments again before playing some music through this receiver.

Beomaster 2400 Type 2902 - Canada Project Volume Muting Circuit When Source Selection Is Made

Every now and then there are some discussions on Beoworld regarding the Beomaster 1900 and 2400 receivers' volume muting during selection of a music source. It is something that I never have really paid much attention to so the discussion caught my interest and I decided to measure the volume muting activity to see it in action.

Here is the Beomaster 1900 and 2400 source selection muting circuit.
One thing to note...I have a few service manuals for both the BM1900 and 2400 but neither show this circuitry for later model PCB3 Volume Control boards. The circuit didn't change but other things on the Volume Control board did. As a result, the later model boards have different reference designators and the location of the components changed some. I found that the new reference designator for the old 3TR6 transistor to be 3TR29. Since the documentation didn't change I will continue referring to the transistor as 3TR6 though.

Here is the trace side of the newer BM1900/2400 Volume Control Board (PCB3).  I have attached three wires (black, yellow and blue) for my oscilloscope probes to connect to.

Here is the source switching volume muting description from the Bang & Olufsen BM1900 service manual.

Now for a look at the signals on the oscilloscope.  For these measurements I am still using the Beomaster 2400 with its original lamps.

The first photo is of the signals being monitored (3TR6-collector, 3TR6-base, 3TR6-emitter and 5R3)...when one source has been selected (Beomaster 2400 "On").

The voltage measurements are pretty much like the service manual states...except the value of 5R3 was expected to be higher than the voltage at 3TR6-base. I will measure these again when I replace the lamps in the PCB 5 lamp board.

Here are the same signals when a source switching event occurs.

As the service manual described, the 3TR6 collector goes to about -12 VDC. That should mute the Beomaster 2400 volume. The duration of the volume muting event I measured is 324msec. The service manual says it should last about 500msec. So I will check the duration after I replace the lamps.

I performed another series of tests using a different PCB 5 lamp board. This board I replaced the source indicator lamps with LEDs and 1KΩ load resistors (to limit the current through the LED to its rated value).  I have used this type of lamp replacement before and it didn't cause any problem with source selection and the display of the selected source. However, I suspect that it does affect the short, ~500msec volume muting circuit.

Sure enough, the LED lamps negate the volume mute circuit during source switching.  The LED assemblies don't use much current compared to the incandescent lamps and the act of switching the selected source doesn't excite the base of 3TR6 to engage the volume muting.

I could experiment with changing the 5R3 resistor and adding some bypass resistors across the LED assemblies to come up with a combination that excites the muting circuit. I would have to be careful because too much current could result in the volume muting always being on.

For now I believe I will stick with incandescent lamps for these source indicator lamps. That way the volume muting during source switching will be functional. This time I will spend a little more time seeing if I can actually hear a difference during the Beomaster 2400 source switching events.

Beomaster 2400 Type 2902 - Canada Project Power Supply & No-Load Current Adjustment

 The hard parts of the Beomaster 2400 are completed. It is time to put the receiver back together so I can do some basic voltage checks.  

First though I need to re-attach the four output amplifier transistors that mount to the Beomaster 2400 heat sinks. I like using the Sil-Pad thermal interface product rather than the white thermal paste in these low profile receivers. The Sil-Pads do a good job and don't have the mess that paste does.

At this point I decided I wanted to check the Beomaster 2400 restoration I had done so far by powering up the Beomaster and performing some basic service manual tests.  In order to do that I re-attached the Filter & Tone Control board along with the Beomaster 2400 lamp boards.

If you think your eyes are playing tricks on you in the above two photos they are not. I didn't really like the new replacement 2C92 2200uF, 25V capacitor for +15VDC in the Beomaster.  The markings on that capacitor were kind of strange. It wasn't as clear as it should be about which end of the axial capacitor was positive and negative. Normally the (-) marking is a distinct arrow pointing at the negative end. This capacitor looks like it points in both directors and the data sheet on it wasn't any help...so I switched it out with a Vishay 2200uF, 25V, axial capacitor that has clear polarity markings.

Plugging in the Beomaster 2400 and turning the master power switch on resulted in the red Standby LED illuminating.

The first test I performed was for the +15VDC supply. I adjusted it to the required +15V per the service manual.

Next were the left and right channel no-load current adjustments.
The space for attaching measurement wires and adjusting the trimmers is very cramped in the Beomaster 2400. My method of performing the adjustment is to mount the measurement wires across the emitter resistors from the bottom of the Beomaster 2400 main board.

That provides the leads I need to attach my DMM for the no-load current measurements.
I adjusted the left channel for the target 12mV across the emitter resistor test points.

Then I adjusted the right channel.

So far everything looks good.

I want to measure the Beomaster 2400 muting circuit while I have the cover off and the original lamps still attached. That is the volume muting that occurs when the music source is switched.  Pressing one of the Beomaster music source switches should send a volume mute signal to the volume control until the source switching is completed.

I soldered on some test wires so I can watch the muting circuit in action with an oscilloscope.

Tomorrow I will measure what goes on with the volume muting circuit.

Beomaster 2400 Type 2902 - Canada Project Capacitor Replacement Part 2

Wrapping up the Beomaster 2400 capacitor replacement is just recapping the tone, filter and FM tuning board. Just eight capacitors on this board. This board also has the three slider controls that I cleaned with Deoxit fader lube. 

Here is the board before replacing the capacitors.

Here is the board after the recap. As with the other boards, the seven electrolytic capacitors were way out of tolerance (50% to 100% off their value). The lone 0.22uF tantalum was fine but in being consistent with my other replacements I replaced it with a WIMA MKS capacitor.

All total, there were 58 electrolytic capacitors and 19 tantalum capacitors replaced in this Beomaster.