My next project -- Heavy

[CurtZHP]

OK, here's another revised schematic.  Added the new features.

The one thing I'm still iffy about is the HPF right before VR1.  I don't think I did it right.

Also, is it worth putting a polarity reverse switch at the output?

 

[abbey road d enfer]

68k is too much for R7, impairs S/N. Divide by 10.
Get rid of R7; it loads the 1st stage so alters gain. Adjust C2 till you're satisfied.
You need a grid leak resistor (already told you)  1 Meg.
Why don't you implement Moamps' excellent suggestion of applying P48 to the center tap of the input xfmr via a single 3.4k?

 

[CurtZHP]

68k is too much for R7, impairs S/N. Divide by 10.
Get rid of R7; it loads the 1st stage so alters gain. Adjust C2 till you're satisfied.
You need a grid leak resistor (already told you)  1 Meg.
Why don't you implement Moamps' excellent suggestion of applying P48 to the center tap of the input xfmr via a single 3.4k?

I went and confused myself again, by having TWO resistors designated R7 in my schematic!


Which R7 should I get rid of?  The one in the HPF?
According to my calculations (which are highly suspect...), the HPF is supposed to give me a roll-off at around 100Hz.
Then replace the 68K on the input to 6.8K?
As for the 1M grid leak resistor, I may have confused grid leak resistor with grid resistor.  Both stages?

Almost there!

 

[emrr]

68k is too much for R7, impairs S/N. Divide by 10.
Get rid of R7; it loads the 1st stage so alters gain. Adjust C2 till you're satisfied.
You need a grid leak resistor (already told you)  1 Meg.
Why don't you implement suggestion of applying P48 to the center tap of the input xfmr via a single 3.4k?

Building out from the above quote:

No 68K, we meant stick the 1M back in. 

The other R7 isn't needed either, right, C2 is the thing.  VR1 IS R7 already, that's the value you calculate against. 

I forgot about the shelf aspect of adjusting V1 cathode cap instead, I'm puzzled by the comment about avoiding electrolytics here, that's always an electrolytic.  Anyway, that could be bigger or smaller as needed to adjust lows if you prefer a shelf option. 

I don't know that there's a problem still with using my 3K4 suggestion before a pad, wasn't thinking about that.  It adds 10% series resistance, probably not a problem in most cases, but a power hungry mic drawing the full spec might cough a bit.  Can't say what the likelihood is.

Listen/measure LS-10 sec ground return connected at V1 cathode instead of ground, likewise with LS-50 Pri ground to V2 cathode.  Listen to C3 at the ground side of LS-50 primary instead of plate side. Another set of tricks from the ancients that's fallen out of style. 

LS-10 sec and LS-50 pri: reverse each of those orientations and measure treble response.  One side will be better than the other usually, sometimes dramatically.  Relates to the orientation of winding capacitance to ground.  Phase to the external world as needed after determination. 

LS-10: if you're super tricky with the tightness of your layout you could consider a series parallel switch on the secondary windings to reduce gain at times when you have a hot mic and a slightly overloading front end, but the pad is too much of a gain hit.  If you at all make that of any wiring length you will hurt treble response and invite noise intrusion, it'd have to be super convenient or clever in layout.  I would not expect using only one secondary winding to be successful, as leaving one floating tends to create resonance effects that bodge up response. 

Put a polarity switch on the output side rather than input side.  Consider phase of your DI input relative to transformer input, it may be reversed from the best orientation of the transformer secondary.  You may have to label your polarity switch as "Positive Polarity Mic / DI".

350VDC is a hell of a lot of B+ for a preamp.  That's 25V higher than the majority of American 10W monitor amps.  Most preamps topped out around 280VDC.  And.  You don't have a bit of filter capacitance, you want at least something between R5/6.  1mfd.  10mfd.  22mfd.  47mfd. 

Choke feed, more output gain.  Look at RCA OP-6 output.  Hammond 156C, about $18.  Buy two of them.  If you pursue that you'll have to size C3 based on measurements, as there's LC resonance to account for, it can be your friend, it can be your enemy.  I'll quote:

With the low cost of the chokes you can use 2 in series. This has 3 things going for it.

1st, the inductance is double so you get 1 more octave on the bottom end of the frequency range.

2nd, because the capacitance of the chokes acts as 2 caps in series you will get better high frequency response. In the preamp I built the 3dB down point went from somewhere near 30K to out past 70K. The single choke had a 4dB dip around 40K. With the pair of chokes the dip was only 2dB. There is a broad very low Q resonance around 60K so with 2 chokes in series it drops 2dB down near 30K, then rises back up gently and finally drops below the -3dB point somewhere above 70K.

3rd, you can mount the chokes on standoffs, bottom to bottom and wire the chokes with one winding backwards. This makes the pair of chokes hum buckers. In my test the pair of chokes picked up 20dB less hum than a single choke.

 

[CurtZHP]

No 68K, we meant stick the 1M back in. 

Got it.  68K is out.  1M is back in.

The other R7 isn't needed either, right, C2 is the thing.  VR1 IS R7 already, that's the value you calculate against. 

R7 is out.  Well....  The resistor is out anyway.  I cleaned up the nomenclature on the schematic to eliminate duplicate numbers, but you get the idea.  I ran the numbers, and what I find is that, depending on the setting of VR1, the roll-off point changes, as one might expect.  The "louder" you go with the HPF switched in, the higher the roll-off point.  Not necessarily a problem, though.  I figure as the level goes up, a little more bass roll off in extreme cases might be desirable.  If not, that's what the switch is for!  Anyway, I sort of averaged things out to determine the value of C2.

Listen/measure LS-10 sec ground return connected at V1 cathode instead of ground, likewise with LS-50 Pri ground to V2 cathode.  Listen to C3 at the ground side of LS-50 primary instead of plate side. Another set of tricks from the ancients that's fallen out of style.

LS-10 sec and LS-50 pri: reverse each of those orientations and measure treble response.  One side will be better than the other usually, sometimes dramatically.  Relates to the orientation of winding capacitance to ground.  Phase to the external world as needed after determination. 

Intriguing.  I'll look into this once I actually start building it.

LS-10: if you're super tricky with the tightness of your layout you could consider a series parallel switch on the secondary windings to reduce gain at times when you have a hot mic and a slightly overloading front end, but the pad is too much of a gain hit.  If you at all make that of any wiring length you will hurt treble response and invite noise intrusion, it'd have to be super convenient or clever in layout.  I would not expect using only one secondary winding to be successful, as leaving one floating tends to create resonance effects that bodge up response. 

Very interesting idea!  Exploring that right now.

Put a polarity switch on the output side rather than input side.  Consider phase of your DI input relative to transformer input, it may be reversed from the best orientation of the transformer secondary.  You may have to label your polarity switch as "Positive Polarity Mic / DI".

That's where I'm planning to put it.

350VDC is a hell of a lot of B+ for a preamp.  That's 25V higher than the majority of American 10W monitor amps.  Most preamps topped out around 280VDC.  And.  You don't have a bit of filter capacitance, you want at least something between R5/6.  1mfd.  10mfd.  22mfd.  47mfd. 

Ignore that voltage.  I've relabeled it "B+" on the schematic.  The 350 figure was a leftover from a previous unrelated design that got copied and pasted into this one.  (Like most of this started out...)

Choke feed, more output gain.  Look at RCA OP-6 output.  Hammond 156C, about $18.  Buy two of them.  If you pursue that you'll have to size C3 based on measurements, as there's LC resonance to account for, it can be your friend, it can be your enemy. 

Looking into this too.

Thanks!

 

[emrr]

C1 is still 10x too large. 

HPF shouldn't see a changing rolloff with VR1 rotation.  That isn't adding up to me.  What freq range?

 

[CurtZHP]

HPF shouldn't see a changing rolloff with VR1 rotation.  That isn't adding up to me.  What freq range?

Neighborhood of 100Hz.

f=1/(2piRC), right?  Could be I need to look over my kids' shoulders when they're doing their algebra...

 

[CurtZHP]

C1 is still 10x too large. 

Another question I've been meaning to ask, hopefully without getting into the weeds....

I've seen a lot of designs that use a 0.1uF cap in that position.  Wouldn't that affect low-end response in some way?  I've understood that the lower a cap's value, the higher the frequency it passes.  I assumed that meant that the lower the value, the less low frequencies pass.  I had read somewhere (probably to do with guitar amps...) that a 0.1uF cap doesn't pass 100% until you get to 3kHz!  Funny thing is, I've run across a coupling cap calculator for guitar amps that shows a 0.1uF cap passing everything down to about 18Hz!

I'm happy to put a 0.1uF cap there.  (I've got a bunch of them!)  But what's the story on that?

 

[abbey road d enfer]

The "louder" you go with the HPF switched in, the higher the roll-off point. 

This shouldn't be; are you sure you have connected the pot correctly? Or have you kept a grid leak resistor?

 

[CurtZHP]

This shouldn't be; are you sure you have connected the pot correctly? Or have you kept a grid leak resistor?

See my Heavy_ver5 schematic.

 

[abbey road d enfer]

I've seen a lot of designs that use a 0.1uF cap in that position.  Wouldn't that affect low-end response in some way?  I've understood that the lower a cap's value, the higher the frequency it passes.  I assumed that meant that the lower the value, the less low frequencies pass. 

That is correct.

  I had read somewhere (probably to do with guitar amps...) that a 0.1uF cap doesn't pass 100% until you get to 3kHz!

That is both wrong AND right! It depends on the load.  Here, with 250kohm load, a 0.1uF introduces a -3dB LF cut-off at 6.3Hz.

  Funny thing is, I've run across a coupling cap calculator for guitar amps that shows a 0.1uF cap passing everything down to about 18Hz!

That would imply a load of about 100kohm...

 

[abbey road d enfer]

See my Heavy_ver5 schematic.

I have seen that it's gone from this schemo, but the actual realization is not always 100% in conformity with plans.
Can you evaluate how much the cut-off frequency changes vs. rotation of the level pot?

 

[CurtZHP]

By the way, the more I look at my latest schematic, the more I think the HPF will do diddly-squat.  Switched in, all I'm doing is putting a .015uF cap in series with the coupling cap of the first stage, which isn't going to make a bit of difference.

 

[CurtZHP]

LS-10: if you're super tricky with the tightness of your layout you could consider a series parallel switch on the secondary windings to reduce gain at times when you have a hot mic and a slightly overloading front end, but the pad is too much of a gain hit.  If you at all make that of any wiring length you will hurt treble response and invite noise intrusion, it'd have to be super convenient or clever in layout.  I would not expect using only one secondary winding to be successful, as leaving one floating tends to create resonance effects that bodge up response. 

emrr....
Just for grins, here's a schematic with what you suggested.

By the way, I looked over an RCA OP-6 schematic and saw the choke you were talking about.  Looks like one side comes off the output of the tube, but the other side looked like it went to the power supply, among other places.  The site where I found the schematic (www.preservationsound.com) showed a small diagram that has the Hammond retrofit between the plate of the tube and B+.

 

[emrr]

yep.  that and the instrument jack have to be a close and short lead as possible.

 

[CurtZHP]

yep.  that and the instrument jack have to be a close and short lead as possible.

Take a second look at my last post.  I went back and edited it to include some other questions.

 

[emrr]

I don't see a question?

 

[CurtZHP]

I don't see a question?

Sorry, didn't word it right.  I was just looking for confirmation that the chokes go in series between the B+ and the plate of the output stage tube, upstream of the coupling cap.

I should probably get the HPF straightened out before I start messing with this anyway.

 

[emrr]

Ah.  Choke replaces the plate B+ feed resistor. 

 

[CurtZHP]

Ah.  Choke replaces the plate B+ feed resistor.

So, something like this...