Miller Effect - Adding WCF buffer overkill?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

BluegrassDan

Well-known member
GDIY Supporter
Joined
Mar 17, 2009
Messages
507
Location
Elizabethton, TN
Hey guys,

Miller Effect has got me down... <commence blues riff>

I have isolated high frequency rolloff at higher gain to the attenuator between gain stages. However, adding a WCF as a buffer after the first gain stage eliminates this, and allows for a lower value attenuator.

This has been confirmed by building the circuit and running sweeps. Miller is a non-issue.

Is adding this additional 12BH7 overkill? That's gonna be a lot more heater current draw and real estate. Is there a better way to buffer between stages to lower impedance and prevent Miller rolloff?

 

Attachments

  • 12BH7 Buffer (Nov 27, 2019).png
    12BH7 Buffer (Nov 27, 2019).png
    170.8 KB · Views: 60
Addng a WCF to a simple tube build is never overkill !

[posted without checking details, but looks OK to me with 10s glance]
 
Can you just move the attenuator to feed the output? The output doesn't have gain so there is no miller effect.

Specifically, connect the attenuator to the anode of the second gain stage and then put 4.7u cap in series with it and ground to preserve the bias voltage. Then wiper to grid of output.
 
I think it is a solution to the problem but probably not the best one. You basically have what I call a Classic topology of fixed gain stages with  attenuators before and between. Your HF roll off problem is caused by too high an output impedance driving the between attenuator which you have fixed by adding a high input impedance, low output impedance stage. My version of the Classic design solves this by combining gain and low output impedance into a single stage by using either an SRPP or mu follower first stage. You could try something similar. Alternatively replace the White follower by a straight forward cathode follower.

Cheers

Ian
 
BluegrassDan said:
What about this...

12ay7 (section 1 gain)
12ay7 (section 2 mu follower)
A mu follower needs two triodes so you could make this the first stage using one whole 12AY7
Low value pot (what value? 10k? 25k?)
Either should be fine driven by a mu fiollower.
12ay7 (2 gain sections with some NF)
12BH7 (WCF)

Ta dah?
You may not need the second 12AY7. You could just set up the 12BH7 as a mu follower which will provide gain as well as the ability to drive the output transformer.

Do you get the impression I like mu followers???

Cheers

Ian
 
Here is another "overkill" three triode schematic. Borrows a bit from the BC3C.

Are there any schematics using the 12BH7 as a regular mu follower? How will the 12BH7 sound as a gain stage compared to the 12AY7? Amplification factor?

 

Attachments

  • Three Triode Preamp, 20191128 (01).png
    Three Triode Preamp, 20191128 (01).png
    174 KB · Views: 24
BluegrassDan said:
Here is another "overkill" three triode schematic. Borrows a bit from the BC3C.

Are there any schematics using the 12BH7 as a regular mu follower? How will the 12BH7 sound as a gain stage compared to the 12AY7? Amplification factor?

I do not know of any schematic using the 12BH7 as a mu follower but Jensen use it as an SRPP in their tube mic pre design:

https://www.jensen-transformers.com/wp-content/uploads/2014/08/as054.pdf

12BH7 has a mu of only 17 (its a bit like a big brother to the 12AU7) compared with a mu of 40 for the 12AY7

Cheers

Ian
 
No rattlers here, but it's been a bumper year for tiger snakes way down here.  [super deadly]

I bumped into one with my lawn mower ...  they're tough as boots, will certainly live another day to harrangue the 'bandicoots'.
 
I also tried two 12AY7s as mu followers. Shows promise as noise is less, but it also has “just barely enough” headroom. Gotta fly out to a few gigs this weekend, but I’ll get back to the bench on Sunday.
 
The latest development is...

12AY7 - mu follower
25k pot
12AY7 - mu follower
12BH7 - WCF (for extra oomph)

No trouble driving the output with this configuration. Ruler flat from 20k to 40k at all gain levels with this configuration.
 
That's a fine line up no question.

AY7 is a great medium-sensitivity performer,  and the BH7 is a fine line driver.

If you really need the gain and you like the complexity, mu-follower  (s) are a great choice.

I like one 'up front' no doubt, I would usually go for something simpler 'in middle' however  [personal choice of course].

Generally the BH7 WCF with a good matching transformer is on my  'top 3'  line amps of all time.
 
BluegrassDan said:
I have isolated high frequency rolloff at higher gain to the attenuator between gain stages.
Can you quantify the problem? A quick simulation shows worst-case about -3dB at 20kHz.
The 200k attenuator, is it 200k in/200k/out 'constant-Z) or more like a potentiometer, with variable output Z?
 

Latest posts

Back
Top