tube buffer impedance helpi

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

mus1k_freak

Well-known member
Joined
Dec 13, 2009
Messages
156
Location
Philadelphia PA
I'm working on a tube buffer to plug into a insert on a mixing board to use as a "plugin" to warm up some of my tracks im using 1/2 a 12ax7 with a b+ of 300v 100k plate resistor and a 1k cathode resistor with a 100uf bypass cap. What impedance do I want for my input and output if I'm.going through an insert? Is 68k to high for the input of the buffer? Right now I'm using a 500k pot for input and one for output to control levels
 
You should have no problem at the input. The input impedance needs to be at least 10K and you easily achieve this with your 500K input pot. You could reduce this to 100K with no problem. 68K  would be fine.

The real difficulty is the output. It needs to be able to drive a 10K input impedance and a 12AX7 with a 100K plate load is not going to do this and achieve a satisfactory signal level at the same time. The output impedance of your 12AX7 as it is is around 50K which is much too high. Perhaps the simplest way to lower the output impedance is to use the other half of the 12AX7 as a cathode follower though I would still be concerned about the ability of it to drive a decent level into a 10K load (it would probably start to clip around +13dBu.

The alternative is to use a different tube operated at a higher current so it has a greater output drive capability. The 12AX7 is surprisingly distortion free whereas the 12AU7 is well known to produce plenty of distortion and can certainly drive the 10K load.

Cheers

Ian
 
In addition to what Ian says, considering your triode stage will be in common cathode mode, there is plenty of gain and level available. If you want to drive modern equipment with 10-20k input impedance, you may just add a 33k resistor in series with the output. Assuming you want +4 at the input of the inserted FX, you would run the 12ax7 at about +12-16dBu, which is not pushing really too hard if you want nice euphonic distortion. I'm not saying it's the best solution, really what Ian suggests (using the 2nd half as a cath follower) is more glorious, but it is certainly worth trying. I've done that on guitar amps and I liked it (for some time ;D).
 
Awesome thanks guys! That's what I suspected with the ax7 having such a high plate resistance. If I go with a au7 instead can I still go with 100k plate resistor and 1k cathode resistor with the 100uf bypass? Also will running pots to attenuate the in and out of the circuit have any negative effect on my impedance? Ide like to use 1/2 a tube so I can run 4 buffer channels with 2 tubes  :)
 
A pot in the output will ruin the impedance. If you use a 500k, you will get a useful range of about 1/10th of a turn. If you use a lower value, like 50k, you will load too much the plate. I suggest you use a 100k pot but not connecting the ground; ideally a RevLog.
 
if you use a 12AU7 you will need to change the plate and cathode resistors to set an suitable operating point. With a 300V supply I would suggest you try something like 22K for the plate load and 620 ohms for the cathode resistor. This should give an anode current in the region of 6.5mA. Connect a 1uF plastic cap to the plate and connect this via a 5K1 resistor to a 10K log pot at the output. This means the smallest load on the tube is around 10K. I did a quick simulation of this circuit which showed that at an output of about +17dBu the 12AU7 in this circuit produces about 2.5% harmonic distortion which is almost entirely second harmonic.


With a pot at the input you can set the drive (and distortion) level and the one at the output will trim the signal to suit the insert input.

Note that this circuit has no power supply noise rejection so you need a really smooth HT supply to avoid hum and noise in the output.

Hope that helps.

Cheers

Ian
 
Great thanks! I threw one together with just some spare parts i had laying around and it sounded great running a vocal track from a session last week through it! once i get the values closer to where they should be (just used whatever i had that was close) I think this will be a cool box! I also have switches for true bypass, a switch to ground the cathode instead of using a resistor for a little bit more gain and darker sound, and a switch that links the output of Channel one to the input of channel two for more drive.  the 1uf cap you have, thats the output coupling cap?, previously i think i was using a .047 sozo cap, using a 1uf cap will work fine in this situation? something like a wima 600v 1uf?  thanks again for all the help!
 
Yes a 1uF 400V part or better will do so the WIMA part you mentioned is fine. You need 1uF to get a reasonable bass response into a 10K load . The -3dB point is about 16Hz with 1uF so a 2.2uF would be even better which gives -3dB at 7Hz.

Cheers

Ian
 

Latest posts

Back
Top