SRPP with buffer - 1MOhm impedance?

[Bertel]

Hi,

I intend using a SRPP circuit based on E182CC with a diamond buffer (to power my 8ohm headphones). The diamond buffer has an input impedance of ~1MOhm. I know that the cathode resistor needs to be matched to the load impedance etc., but general question:

Does anyone have experience in operating an E182CC-based SRPP with such high load impedance, does the SRPP "like" to play against such a load, what is important to take care of in such a setup?

Many thanks!
Robert

[Merlinb]

Originally Posted by Bertel

Hi,

I intend using a SRPP circuit based on E182CC with a diamond buffer (to power my 8ohm headphones). The diamond buffer has an input impedance of ~1MOhm. I know that the cathode resistor needs to be matched to the load impedance etc., but general question:

Does anyone have experience in operating an E182CC-based SRPP with such high load impedance, does the SRPP "like" to play against such a load, what is important to take care of in such a setup?

Many thanks!
Robert

1Meg is so high that it's practically no load at all! You would probably be better to arrange the stage as a half-mu amplifier, rather than an SRPP. (e.g., the first stage of the Aikido http://www.tubecad.com/2004/AikidoAmp5.gif). After all, the SRPP is a power amplifier, whereas all you need is voltage gain.

If you actually try to optimise the SRPP for a 1Meg load you will find yourself calculating cathode resistors in the region of 80k! As a rule, the SRPP doesn't like loads that are much greater than the anode resistance of the triode being used. Or perhaps I should say that with large loads, it stops being an SRPP and becomes an actively loaded triode, and the idea of optimised cathode resistors no longer applies.

[Bertel]

Hi Merlin,

many thanks for your reply!

Firstly, thanks for pointing out to me the rule that the SRPP doesn't like loads that are much greater than the anode resistance of the triode being used. You mean the resistor between the anode of the lower triode and the cathode of the upper triode? This is currently at 200ohm (there is no resistor to the anode of the upper triode).

Secondly, a bit more information on the intended usage: This SRPP circuit is the "I/V stage" of my DAC chip. In fact it is a current output DAC anyway, so the voltage gain I get from the SRPP is absolutely fine. I try to drive my headphones with it, and since they have an impedance of 5-14ohm (depending on frequency) I thought it would be a good idea to use a buffer after the SRPP to just provide enough current to the headphones (voltage is enough).

May I ask what you would recommend in this case? It would not be a good idea to drive such low impedance headphones directly with the SRPP, right...?

Many thanks
Robert

[Geek]

You want a mu-follower/stage, they're practically immune to loads above a certain amount.

That light of a load, pretty any dual-triode tube will do. There's also hybrids: http://www.diyaudio.com/forums/blogs...-follower.html

How much voltage gain do you need?

Cheers!

[revintage]

There is no need whatsoever to use a "follower" in your configuration. Use the "follower" that Geek suggests but take the signal from the ANODE instead. The sand then mimics a anode-choke as it now is a gyrator that also controls the anodevoltage. This way you will also get a spare E182CC.

[Bertel]

Hi Gregg and Lars,

many thanks for helping me with this!

I think to fully understand this right I need to take a step back and would like to once more look at the full picture please:

I do not think that I need any voltage gain at all. My headphones are Ultimate Ears UE-10pro in-ear monitors with an impedance of 5-14ohm, depending on frequency. Probably due to being in-ear headphones, they have a high efficiency and play pretty loud. I do not have a scope, but when I put my multimeter over gnd and V+ on either left or right headphone output channel of my current amp, I measure no more than 35mV at listening levels that I would not want to exceed.

For this reason I thought that I would not need voltage gain but would be best served with a power amplifier that focusses on current, not voltage. Would you agree to this?

I intended to use my E182CC SRPP stage for this, but have been told that this is not a good idea with such a low load impedance of ~8ohm. Would you confirm this?

So my idea was to use a buffer "in between" that could supply the headphones with the required current, but now the buffer's high input load impedance is the problem...

Thank you,
Robert

[revintage]

Is the diamondbuffer such a bad idea?

With the high sensitivity one might even get away with running the phones directly from your current DAC with the phones as I/V resistor !

[Bertel]

Hi Lars,

I like the idea of "nothing in between" DAC and headphones a lot - unfortunately the DAC (it is the Wolfson WM8741) is a voltage output DAC (I wrongly wrote "current out" before), so I'm quite convinced I need a current amplifier in between, right?

Thanks,
Robert

[revintage]

Ah,
Then you need balanced to unbalanced unity gain to start with.

I´d suggest a 1:1, (10k:10k transformer with primary termination) followed by volumecontrol then a onestage triodeamp( 1/2 182CC) with output transformer ca 5k:8 ohm. Otherwise your diamond buffer could be used.

An E182CC White-follower after the input transformer might also work.

[Bertel]

Lars,

the UE-10pro have no common GND line but a dedicated pair of GND and V+ to either side. Would it help if the whole amp stage (maybe even all the way including cabling) was to be designed balanced?

Thanks
Robert