[Sarchi]

I got the TO-220 package IC's, thanks to Marshall, as well as heatsinks. I have a working 24V Aikido set up as a linestage preamp, I want to convert it to headphone duty with the BUF634's. This looks easy, but can someone walk me through the connections?

I'm refering to Broskie's blog post 0067. Oops.. I need to go out and get a stereo 1/4" jack, also I don't have the extra cap and resistor, how crucial are they?

I plan to drive low-ish impedance 'phones with this (ATH-W1000).

thanks in advance.

 

[kvant]

Quote:

...also I don't have the extra cap and resistor, how crucial are they?

That depends, which cap and resistor you mean. Those 1k and 3900pF on the input of the buffer could probably be omitted. After all, this BUF634 application is analogous to the original Millett Hybrid, where these components were not used. On the other hand, the cap on the output is crucial, since the output sits on cca 12V of DC offset. With the value of this cap you could go down to 470uF (but probably not much lower, since those W1000 are only 40 ohm, if I am not mistaken). Whether you need the 1uF (of other value) bypass is quite another question. The 10k resistor on the other side of the output cap prevents the output from wandering away from 0, when no phones are connected.

 

[Sarchi]

Quote:

Originally Posted by kvant /img/forum/go_quote.gif That depends, which cap and resistor you mean. Those 1k and 3900pF on the input of the buffer could probably be omitted. After all, this BUF634 application is analogous to the original Millett Hybrid, where these components were not used. On the other hand, the cap on the output is crucial, since the output sits on cca 12V of DC offset. With the value of this cap you could go down to 470uF (but probably not much lower, since those W1000 are only 40 ohm, if I am not mistaken). Whether you need the 1uF (of other value) bypass is quite another question. The 10k resistor on the other side of the output cap prevents the output from wandering away from 0, when no phones are connected.

Thanks a lot, that is really helpful.

For some reason I had a brain fart and didn't really notice the output side; JB didn't mention it, and I have schematic-blindness. So I guess I'm shopping for some different coupling caps again. Makes sense.

Actually I have a couple of 470 MKC Philips here, maybe I'll give these a try.

 

[Sarchi]

More questions, for those familiar with the circuit.

1. My C3/C4 are all jumpers, is this still ok for the headphone version of the Aikido?

2. Wondering what is the best way to put the buffers into the circuit without permanently altering the board, i.e. cutting traces. Can I just run the buffer's input from the R14 lead, and the buffer's output to my C2 cap? (I have these MKC's which fit right in on C2)

3. If I wanted an "ultimate" coupling cap, what should I look at? I'm assuming kvant meant 470pF, not uF.

 

[fran]

I think your C3 and C4 should be OK as is.

When I thought about doing this I was going to wire up the 634s on a separate breadboard, or even make a little PCB for them and the caps.

You'd want to check, but I think it is uF! But I think you can use an electrolytic bypassed with a film.

Fran

 

[Sarchi]

OK, thanks Fran.....I think I get it. The Millett Hybrid schematic shows the BUF634 output going to C7 (>100uF electrolytic), bypassed by C4 (0.22uF film). I'll work with that in mind.

The MKC's I have are labeled as 470n. The voltage is way overrated for this...630V. So is that rating the same as 0.47uF? Sorry for the dumb questions...

I guess I want some Black Gate N-series 470uF's then....will that work well with the film bypass caps I have? Or can someone suggest other bypass caps. I can order from Pcx, they're just down the road from me. The load is 40 Ohms (ATH-W1000).

 

[dsavitsk]

If you are going to bother building an Aikido just for voltage gain, then you might as well use a decent buffer with it. The buf634 sounds terrible and will kill any goodness the front half might give you. Get in on the JISBOS group buy at Headwize for a better buffer. Then you can use a very small high grade coupling cap between the Aikido and the buffer, build a virtual ground for the buffer, and do away with the output cap which is likely to destroy any goodness left over after the 634 is done with the signal.

If you don't want to build a virtual ground, then do something like the following replacing everything between the two caps with your Aikido circuit. Essentially, the -V will become the ground for the tube section which is AC coupled to the rest of the world. You would want +/-12V supplies here, which the Aikido would see as +24V and ground.

 

[dsavitsk]

Here's what it would look like - you might want to increase that first 100K resistor to 250K or 500K

 

[Sarchi]

Sounds interesting. I have no clue how to do a virtual ground though.

 

[fran]

Voltage rating is voltage rating, no matter what the capacitance is.

So your 630v rated caps are way over rated in terms of voltage but as nF (nano) are 1000 times smaller than uF(micro) then your capacitance is 1000 times too small. In other words 470nF is 0.470uF.

About here is where you realise that getting a decent cap of 470uF is going to be expensive! Maybe some russian PIO caps or something off ebay might do.

As far as dsavitsk goes - I just don't know about the sonic qualities of the 634. However I recognise Dsavitsk from other forums and the fella knows what he's talking about! It would wise to check out that group buy.

Fran