Mic Pre

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
clintrubber said:
Don't recall which mods he did, possibly using an input-TX. 
This seems to me inescapable for a mic preamp.

Note there's a low-noise mod suggested here, which may come in handy for mic-preamp use:

http://w140.com/brophy_122_mod.pdf
Very interesting; it's the first time I see decreasing heater voltage as a significant means of reducing input noise whilst maintaining transconductance.
 
abbey road d enfer said:
Very interesting; it's the first time I see decreasing heater voltage as a significant means of reducing input noise whilst maintaining transconductance.

Same here, not read such a thing before. Makes one wonder why it's not used more often. Clashes with the recommendation to avoid heater voltage deviations of say more than +/-10% though (in view of life time and/or shifting parameters), perhaps that why it's not considered more often. But hey, if it gives lower noise then simply replace the tube more often. We change guitar strings as well.
 
Step up to grid transformer/resistor  Id have assumed would be a first step to modding it for dynamic mics at least ,
That being said would it be possible to match a 48v condenser mic directly to a push pull valve input stage , re-jig the ht+/- to show 48v+ at the grids to energise it ,even if it was only usable on high level sources.

What I find hard to visualise is a resistor with + volts at the top and - volts the other end , so at some point along the resistor is 0 volts ,yet ground and chassis is also 0 volts yet not connected  :eek:

Many thanks also to Ian and Abbey for the insights on feedback .

He didnt mention AC or DC supply in the modifications , what was the original type 125 I wonder?





 
That wide body rack mount RM122 would look nice on top of  a Tele LA-2a , LF filters are good as they are , HF cut to 50,250,1000hz makes no sense for audio ,repurposed to more usable frequencies in the khz band is easy to do  .
122 output stage is only good for hi(ish)-z input though ,


 
Tubetec said:
Step up to grid transformer/resistor  Id have assumed would be a first step to modding it for dynamic mics at least ,
That being said would it be possible to match a 48v condenser mic directly to a push pull valve input stage , re-jig the ht+/- to show 48v+ at the grids to energise it ,even if it was only usable on high level sources.
That is perfectly doable. It is also doable with solid-state circuitry.  It would just take lifting the +/-15V rails to +33/+63; not difficult but adds complexity. Then the reason why wanting to do that is getting rid of the input capacitors, but the issue is the output signal is then lifted by 48V, and needs to be shifted down to 0V, which takes at least one of the dreaded capacitors. Or something else (e.g. xfmr-coupled  isolator).
 
abbey road d enfer said:
Very interesting; it's the first time I see decreasing heater voltage as a significant means of reducing input noise whilst maintaining transconductance.

From my experience, 12ax7 can't work at 3V for heater voltage.
Also it is questionable does the heater voltage stay at -6V when the heater current is smaller. It is much higher, IMO, because of installed power resistor across the pass transistor.  They stated BTW  the minimum heater current of 0.7A for TEK125 power supply.  If only one "modified" preamplifier is connected to it, the power supply is out of regulation for filament part, IMO.
Schematic of power supply is here
http://bama.edebris.com/manuals/tek/122_125/
 
moamps said:
From my experience, 12ax7 can't work at 3V for heater voltage.
Although I never tried it, I tend to believe that.

Also it is questionable does the heater voltage stay at -6V when the heater current is smaller. It is much higher, IMO, because of installed power resistor across the pass transistor.
Not knowing the unregulated voltage value does no help analysis but no doubt the 9 ohm resistor takes over regulation. Then the remaining tube in the third stage, that is not series connected get excessive voltage.
I don't think Tektronix would have approved of the mod.
 
I have a regulated LT supply on an EL84 based line driver , I did notice hum drops disproportionately as you reduce heater volts , and that even holds true for a relatively quiet dc supply  ,everything else apart from gain and anode volts drops too .

I'd be of the mind that if we were to reduce heater volts ,we should also de-rate the anode voltage , in other words if a valve is operating at a fraction of its maximum anode current ,were quite safe to shave a volt or maybe more off the heater supply ,if your running a tube up towards the limits of its disipation  dont be mean with heater volts .

3v on an ECC83 heater  ;) best of good luck with that unless you want to run it with a 470k  load and starvation anode volts well  <100. 
Its often advisable to check back on the correspondence page letters of advice the following month ,as retractions ,corrections and additions are made quite often .  ;D
 
moamps said:
From my experience, 12ax7 can't work at 3V for heater voltage.

I can attest from personal experience that it does. A long time ago I made a long series of measurements of double triode tube noise when configured as a mu follower. When I first tested a 12AX7 I accidentally wired the 6V heater supply across the 12V heater pins so each half was working with 3V heaters. It did work but the gain swas eriously reduced. I forget by how much because I soon found the fault and fixed it.

Cheers

Ian
 
abbey road d enfer said:
This seems to me inescapable for a mic preamp.
Very interesting; it's the first time I see decreasing heater voltage as a significant means of reducing input noise whilst maintaining transconductance.

I do not recall the article saying transconductance was maintained. Overall gain increase because the second stage 12AU7 (u=17) was replaced with a 12AX7 (u=100) but I do not see any mention of transconductance.

Cheers

Ian
 
A valve will work at 50% heater volts , but the curve wont relate well to how it behaves at 6.3 surely?
 
ruffrecords said:
I do not recall the article saying transconductance was maintained. Overall gain increase because the second stage 12AU7 (u=17) was replaced with a 12AX7 (u=100) but I do not see any mention of transconductance.
I should have writ "maintaining some transconductance". Indeed, I would not expect total conservation of electron flow.
According to the paper, overall gain increased by a factor 2.8, when the 12AU7 to 12AX7 would have accounted for factor 5. That means that Gm decreased by only about 30-40%. That is very surprizing to me.
 
ruffrecords said:
I can attest from personal experience that it does. A long time ago I made a long series of measurements of double triode tube noise when configured as a mu follower. When I first tested a 12AX7 I accidentally wired the 6V heater supply across the 12V heater pins so each half was working with 3V heaters. It did work but the gain swas eriously reduced. I forget by how much because I soon found the fault and fixed it.

Isn't that just confirmation of what I said? Come on.

12ax7 can work from, let say, 5V because it has small heating element and anode current is low.

However, someone can try cold emission with it and will get some anode current. There are always some desperate free electrons  around.  But it isn't 12AX7 anymore.

As I said,  I believe the heating voltage in question wasn't drop to 3V because of the lousy filament regulator and gain of tubes remain reasonably high.
 
No need for heated (pun intended) arguments here. The issue there is defining "work". A 12AX7 with 12V plate voltage is not a 12AX7 anymore, or is it? A number of (questionable) products do that, though. I think a more scientific approach is desirable here; anybody with a small jig and a variable PSU could check.
 
abbey road d enfer said:
No need for heated (pun intended) arguments here. The issue there is defining "work". A 12AX7 with 12V plate voltage is not a 12AX7 anymore, or is it? A number of (questionable) products do that, though. I think a more scientific approach is desirable here; anybody with a small jig and a variable PSU could check.
This is the brewery... better to argue about tubes than the typical subjects.  ::)

JR
 
I think a few behringer tube products ran 24 or 36 volts ht , and reduced heaters on 12ax7's , it adds the tube processed signal in parralel with the op amps which do all the work ,
 
There have been any number of efx SKUs that ran tubes at low voltage just so they could say the audio path included a tube, and avoid the obvious cost of HV PS and components. How they sounded was not paramount, what the FAB (features,advantages,benefits) card said at POS (point of sale) was....  ::)

JR
 
Back
Top