DIY: Summer Projects

[solderdude]

The K712 filter is VERY subtle in action but audible. The same is true for the HD58X.

Some folks prefer the sound as it is, I prefer them with the filter.

 

[Silou]

K712 works as well. I will give it a listen in the evening.

 

[Silou]

Is it possible to estimate or even calculate the output power of the DualO2 yet? It is probably not as easy as to just multiply the output power of the normal O2.

This enclosure is probably perfect for the project. Two rotary controls for volume and filter selection and three LEDs to indicate which filter is currently in use. Quite expensive though.

XLR combo sockets are probably not bad if an unbalanced source is connected.

 

[solderdude]

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

Running on 9V batteries = max Uout: 10V
600 Ohm: 0.16W
300 Ohm: 0.3W
150 Ohm: 0.65W
60 Ohm: 1.1W
32 Ohm: 0.6W

K712 = 0.2W rated (60 Ohm)
HD58X = 0.5W rated (150 Ohm)
Amiron Home = 0.25W rated (250 Ohm)

 

[Silou]

I can see that you multiplied 600,300,150 Ohm with 4 but why is the output into 60 and 30 Ohm not increased?

 

[solderdude]

Because the output op-amps have current limits and as the outputs are in series the current won't increase, only the voltage swing can (and will) as long as the max current is not reached.
A 'normal' amplifier is a voltage and current limited voltage source.

I used available data (power rating in 32 Ohm as current is limited there) and max output voltage of the O2 (7.2V, mains fed)
So the power numbers above may differ in reality a bit ... but not very much.

 

[Silou]

Can this be 'fixed' by using different opamps?

 

[solderdude]

Yes you could increase the output current capabilities using different opamps.
You could also increase the max output power for higher impedances by running the opamps on +/- 18V instead of +/-12V (when the opamps used support that.
In such a case I would suggest to use an entirely different board, one that already is balanced and can run very different opamps and output devices.

 

[Silou]

In such a case I would suggest to use an entirely different board, one that already is balanced and can run very different opamps and output devices.

Can you recommend a board design with available PCBs? I do not need tons of Power but I guess more Power is not bad either.

 

[solderdude]

The setup you want to build is already exceeding the max power ratings of your headphones.

Unless you want it to power inefficient low impedance planars like the HE-6, Audeze Sine, MyST Izo-OS, HE500, HE4 I think the bridged O2 concept will be fine.
Different output devices can also help a bit.

 

[Silou]

Unless you want it to power inefficient low impedance planars like the HE-6, Audeze Sine, MyST Izo-OS, HE500, HE4 I think the bridged O2 concept will be fine.
Different output devices can also help a bit.

A planar headphone is definitely planned. Based on your measurements most planars are insensitive to higher output impedances.
Is it possible to put an adapter between the headphone and the amplifier that "pretends" that the headphone is of higher impedance to get a higher output?


"For those who want the exact numbers, or want to change the default gain settings, on AC power the Maximum Gain = 7 / Vin(max) and on (low) batteries it’s 4.5 / Vin(max)."

That means that the Max Gain is 3.6X, because the voltage is doubled in a balanced setup. Am I missing something or is that correct? I calculated 1.75X in a previous post but did not pay attention to the doubled voltage.

 

[solderdude]

An extra resistor would only be recommended for the likes of Verum 1. As that one has a high efficiency it does not need that much voltage so the extra resistor trick works great in that case.

Also something like the AEON could use such an extra resistor but that one too is high efficiency so can also easily be driven.

The total gain increases (by 6dB) but the clipping point of the input stage remains at the same point so those numbers remain the same.
You could lower the gain a bit and increase the max input voltage this way (make it half as sensitive) but that would only be needed if you are always using the lowest part of the volpot range.

As an alternative you could build Pete Millets 'Butte' but increase the operating voltage (requires some different parts)
You can run it on +/- 24V (gives same output voltage as bridged O2) but is can supply 40% more output current and does not have to be balanced.
A single board can do this. This means about double the output power of the O2 for low impedances and the same output power for high impedances as the balanced O2.
For gains below 5x it is better to use the OPA551.

Of course, when driving high efficiency IEM's the O2 will be far less noisy.

With some tinkering (a home made DIP8 converter) you can also use the OPA551 instead of the 4556 and double the output power of the O2 bridged amp concept.

 

[Silou]

With some tinkering (a home made DIP8 converter) you can also use the OPA551 instead of the 4556 and double the output power of the O2 bridged amp concept.

I think this is a great idea, but I do not get why a converter is needed. I thought I read that the O2 uses DIP8 packages and the OPA551 is available in a DIP8 version aswell.
Edit: OPA551 is a PDIP8

 

[solderdude]

The DIP-8 package is the same.
The in- out-put pinning is not.
The OPA551 is a single op-amp (with extra in/outputs for specific functionality) where the ones used in the O2 are dual op-amps.
yet, the OPA551 can supply more current than the 2 op-amps combined in the same package.
Under heavy load the DIP package can get quite hot (thermal protection inside)

 

[Silou]

Is the pin layout a matter of manufacturer?
Seems like nwavguy tested the OPA551:

• TI Burr Brown OPA551/2 – These single op amps were my first choice as they look good on paper, are rated for 200 mA of current, have high open loop gain, and even some cool features like thermal shutdown. In testing, however, they proved to be something like the AD8397—not quite as bad but still high strung. Admittedly my prototype board wasn’t optimal but the OPA551 didn’t like reactive loads without using output inductors or substantial series resistance. You can get clever with the feedback to presumably compensate for capacitive loads but it kind of works against the OSFA mantra as it’s hard to optimize it for all possible loads without using an output inductor. The very high current limit of 380 mA is also less than optimal. Finally, they’re relatively expensive at $5 each and you need two of them.

 

[solderdude]

A matter of functionality.
There are many brands that use similar pinning for single and dual opamps in DIP-8 or SMD versions.
Some devices (single op-amp) can have adjustment, compensation or other function pins assigned to certain pins.

Most dual opamps are pin compatible but not necessarily function compatible because of certain aspects (minimum gain, input or output stage or power supply voltages to name but a few of the many possible deviations.
That's why sometimes 'brainless op-amp swapping' can lead to unexpected or different results and is something I generally won't recommend.
One needs to know if certain parts are interchangeable in certain applications.

PDIP stands for PlasticDIP which most are but CeramicDIP also exists.

 

[solderdude]

Seems like nwavguy tested the OPA551:

Yes he tested it but could not possibly have just fitted it in the DIP on the PCB.
It simply cannot possibly work that way as the pinning is vastly different.

I have experimented with using OPA551 in parallel (like the O2 does) and could not get it to work properly, the 380mA capabilities point in that direction. Works great when only a single opamp is used and using local feedback without a resistor in the output/feedback path.

 

[Silou]

I have experimented with using OPA551 in parallel (like the O2 does) and could not get it to work properly, the 380mA capabilities point in that direction. Works great when only a single opamp is used and using local feedback without a resistor in the output/feedback path.

So you would not recommend the OPA551 for the balanced O2 setup.

 

[solderdude]

I would leave the O2 'original' as you do not seem to need the extra current... plus it is cheaper to use 4556.
Will be fun to make a balanced O2 as well.

 

[Silou]

Nwavguy commented on the bridged idea of his O2. What do you think about his comment?