DIY: Summer Projects

[Silou]

It works well Are the LEDs supposed to be brighter when connected to ahigher voltage? I think I saw no difference between 12V and 20V

 

[solderdude]

There is most likely a current source which always ensures the same brigthness.
The fact that is was spec'ed 4.2mA already implies that there isn't going to come much brightness from it.
I would expect bright LEDs at 12V to have a rating of at least 50mA or so.

 

[Silou]

I tested it again with no other lighting in the room and it gets brighter with higher voltage. I probably could not see it because the increase in brightness is not enormous and the room was already lighted well by the sun.

 

[solderdude]

You can increase the voltage to max 24V (so connect it between the +12V and -12V) in which case the current will have increased to 20mA.
That is as bright as it will go.
It is intended to be an indicator light not for illumination.

 

[Silou]

I don't think you are planning to light up the room
They will probably between 2.2k and 100k.
I would suggest to buy a 1k resistor + 100k potmeter and visually tune each LED.

I know what you meant now The white LED is so bright that it actually hurt my eyes
The potentiometer "trick" works perfect. Even the 2mA LED would be too bright without the pot.
The yellow LED matches the signal rings colour. Looks really good

This is the test setup Selection switch works well aswell

 

[Silou]

I did not really find a way to properly mount the filters into the case, so I soldered them onto little PCBs. The idea was obviously stolen from the Kameleon amplifier

 

[solderdude]

Have you tried running the LED ring on 24V yet ?

 

[Silou]

Have you tried running the LED ring on 24V yet ?

I have, but only with a seperate power supply. I will try it with the board once the front panel is completely machined.

 

[Silou]

The front panel is done now. It was really not that easy. As you can see there is almost no room to mount the audio connectors. The holes are both 24mm wide. I wanted to mount them inside, because I think that looks better, but that is certainly not possible with a 8mm thick panel. We used a M3 thread cutter so we do not need nut screws to secure the connectors. Next problem was the power switch. It could not be mounted in the center, because it would have interfered with the mounting stuff of the side panels. We mounted it into the center of the left part including the part where the panel thickness is lowered. The signal rings fit perfectly into the holes. We just needed to add 5,5mm holes for the cable feed through. We did this by hand, because it is not visible and did not have to be perfectly accurate. The Alps pot fits perfect without any modifications to the panel. The filter selector did not though. We had to increase the middle mounting hole to 10mm.

Now it is time to solder again
Which cable connections inside the case require shielded cables and which do not?

 

[solderdude]

There is no transformer in the case so there is no real need for screened cables.
The incoming AC that runs to the PCB I would recommend to twist.
The ground wire from the TRS to the positive PCB ground should be thick (say 1mm sq to 1.5mm sq)

wires to and from the power switch = min 1mmsq.

When you have screened cable around you can use it for the audio input wiring and potmeter.
For the latter it is not really needed.

 

[Silou]

How can I wire the Power and Gain switch correctly? I could not find any pin labeling or other indicators. The Gain switch has 6 pins which are labeled 1-6. The Power switch has 6 contacts aswell and is symmetricaly labeled. 2x COM 2x NC 2x NO. I think you mentioned that only three contacts of the Power switch need to be connected to the board. COM is ground, NC is close and NO is open?

 

[solderdude]

Try to find the data sheets for the gain switch.
The manufacturer usually supplies this data.

NC is Normally Closed and NO = Normally Open.
For the power switch you thus use COM and NO.
Normally is the not activated position.

In the O2 schematic:
1= N.C. (+12V)
2= COM (+12V)
3= N.O. (+12V)

4= N.C. (-12V)
5= COM (-12V)
6= N.O. (-12V)

The LED ring you connect between 3 (red wire) and 6 (black wire)

 

[Silou]

This is the Gain switch. There is a drawing in the description, but I do not really understand it.

 

[solderdude]

The middle contacts are the COM.
These go to ground (pin 2 and 5 of the O2)
The ones on 1 side go to pin 1 and 4
The other gain position to pine 3 and 6 respectively.

 

[Silou]

For the power switch you thus use COM and NO.
Normally is the not activated position.

In the O2 schematic:
1= N.C. (+12V)
2= COM (+12V)
3= N.O. (+12V)

4= N.C. (-12V)
5= COM (-12V)
6= N.O. (-12V)

The LED ring you connect between 3 (red wire) and 6 (black wire)

So Pin 1 and 4 are not going to be connected? What about the three indicator LEDs?

 

[solderdude]

correct, 1 and 4 are not connected for the power switch.

The 3 indicator LED I would power from C8.
Each LED will be drawing less than 1mA most likely and only 1 will be active.

 

[Silou]

Running on mains = max Uout: 14.4V
600 Ohm: 0.35W
300 Ohm: 0.7W
150 Ohm: 1.3W
60 Ohm: 1.1W
32 Ohm: 0.6W

One more question to the calculated numbers.
The O2 is limited at ~200mA so the max. power it should be able to output at 14,4V is 2,88W into 72 Ohm.
At 200mA and 12V the amp should in my theory output 2,4W into 60 Ohm and 1,28W into 32 Ohm at 6,4V.
Even at only 150mA the amp should output 1,35W into 60 Ohm at 9V and 0,72W into 32 Ohm at 4,8V.

32 Ohm 0,6W -> 0,137mA 4,384V
60 Ohm 1,1W -> 0,137mA 8,220V

You can find these numbers everywhere, but why is the amp not able to output more current? I am just curious what the "problem" is here.

Edit: I think I found the mistake myself. These numbers are correct, but I have to divide them by two, because two channels need to be powered. I am not sure though.

Edit 2: What do you think about the O2 booster board? Sounds interesting, but it is most likely not plug and play like on a normal O2.

 

[solderdude]

Current is limited by the used op-amps.
For balanced these are in series.
So while the output voltage doubles (when the max current is not reached) max current is not.

The spec sheet shows the current limit for the positive signal = 100mA (knee at 70mA so the opamp is specified at +/-70mA)
The Negative side of the opamp limits at 200mA.
This is per opamp. Since 2 in parallel the current doubles. So 200mA max on the plus side and 400mA on the negative side.
But for a sinewave we don't want clipping asymmetric clipping so 200mA is the max, otherwise distortion goes through the roof.
That is the reason the output power drops dramatically once the current limit is reached.
My earlier calculations were based on the specified +/-70mA per opamp so 140mA for O2

For the balanced O2, when 200mA is the output current, the max. output powers will be:
600 Ohm = 0.35W (current limit not reached so max output voltage)
300 Ohm = 0.7W (current limit not reached so max output voltage)
150 Ohm = 1.3W (current limit not reached so max output voltage)
So far so good calculations are straight forward as enough current is available so voltage limit determines power.
At 72 Ohm one would expect 2.9W based on the 200mA output but assuming the output voltage doesn't drop.

At 100mA (per opamp) the positive side only puts out +4.5V, this is also seen on NwAvGuy website as well as in the datasheet.
The negative side, however, is -9.5V so 14V peak = 10Vrms (when bridged) and not the expected 14V so when power is calculated 10V@72 = 1.4W
At 60 Ohm the same happens the current is limited at 200mA = 2.4W BUT the max output voltage will be 10V so power will be limited to 1.6W in 60 Ohm.
At 32 Ohm = the current is limited to 200mA = 1.3W because the positive side, which is in series with the negative board simply can't deliver more.

 

[Silou]

In the O2 schematic:
1= N.C. (+12V)
2= COM (+12V)
3= N.O. (+12V)

4= N.C. (-12V)
5= COM (-12V)
6= N.O. (-12V)

The LED ring you connect between 3 (red wire) and 6 (black wire)

Can I solder the LED rings directly to the power switch?

 

[solderdude]

Yes.