Bass pre-amp

[RPmanen]

Hi there,

I've been thinking about building a bass pre-amp to amplify my signal to line level . I am going to use it for recording, going into an EMU 1820.

I've found the Alembic F2-B schem at Geofex, and would love to build one. The only thing is, I can remember reading about a problem with this schematic.

Would the Alembic be powerfull enough to feed into my soundcard (line level), and is the schem correct? Also, if any bass player here has built a nicer pre- feel free to say so, as long as you post the schematics as well.

Link to schem: http://www.geofex.com/FX_images/alembpre.gif

Rob

[Chadwick]

very good start

The audio signal looks like a great start though you'll get plenty of voltage gain but possibly not enough current gain.

The filiment is suspect, it looks like they are rectifying it and maybe floating it but with such a small unit you could just regulate it at 12v and eliminate a lot of hum.

I tried something similar and built the power too close to the tubes and it caused hum.. I just prototyped a walwart power supply using a 500ma 18vac supply and a cascade rectifier to get 120vdc plate voltage.

I used a 12vdc regulator for the filiment and haven't yet ordered the 120vdc voltage regulator for the plated but the hum was a lot better.

I think my next proto will use a cathode follower and only one stage of gain since mine seemed to not drive the next thing in the pathway very well.

[RPmanen]

Yeah I was looking at the powersupply, I thought I might use a 12V AC wallwart, and split it, one way rectifying it and regulate 12v for the heater, the other into a transformer 12 to 220 or 230VAC, rectify for high B+ voltage. should be around 300VDC right?

Rob

[Chadwick]

math

If my math is correct rectified DC is 1.4 times the ac.

If you get to 150 and above your golden though.

I'll try and post the cascade rectifier here... I think it sucks a lot of current but the plate doesn't need hardly any so its okay..

Ya the transformer is a good idea too but I would worry about EMF hum introduced into the circuit. I did the cascade for that reason but its not perfected so anything could work. If the unit was in a larger metal and you mount the tranny on top, no problems there I'll bet.

The cascade...


Thanks to Radio Shack's 'Basic Electronics' for that..

[Rob Mercure]

Chad,

The "cascade" is more commony referred to as a "multiplier" (or simply a doubler, tripler, quadrupler, etc., for a few stages) - where in the hades RS got "cascade" I dunno but in close to 40 years of electronics I've not encountered the term. Like I suggested on another post it makes it easier on others who are less experienced to adopt the commonest terms (hey, I had to learn to say "long tailed pair" instead of "cathode coupled grounded grid" PI so we can all learn). Oh, and the circuit can't "suck" current but instead you have to divide the original available maximum current by the number of multiplier stages to derive resulting available current - TANSTAAFL ("there ain't no such thing as a free lunch" - R. Heinlein '"The Moon is a Harsh Mistress"). And while peak voltage is 1.414...times RMS - and a good thing to remember when sizing capacitors - it drops under load. But in this case with such a light B load you may wind up with something pretty close to peak.

The "Alembic" preamp screams "Fender" all through the design so if it ain't whatcha want reviewing Fender schemos may provide fertile ideas.

And I wouldn't worry about any extra EMF in this case unless one is truly trying to cram too much in too little space (guilty as charged, yer honor) - while any capacitive/electrostatic coupling is a function somewhat of voltage levels what we usually fight with trannies is the electromagnetic field which is a function of current and number winding turns (and if you use a sausage for your core you get a "turn for the wurst" <grin/groan>) and the heaviest current flow is in the heaters which you "wind" up with no matter how you derive your quite low current B+. And the heater is neither rectified nor floated - the rectifier "past" the heater connection just provides for an LED pilot light (which might just "rectify itself" with careful choice) while there is a 250 ohm pot with its wiper grounded to reference the heaters - but in practice one might have to still swap the tranny heater leads to find the least hum connection, I dunno.

Rob

[David Schwab]

The "Alembic" preamp screams "Fender" all through the design...

That's exactly what it is. They have even said they took the front end of a Fender amp and put it in a rack enclosure.

Would the Alembic be powerfull enough to feed into my soundcard (line level), and is the schem correct?

I can't say if the schematic is correct, but you can drive a power amp with the F2-B, so I don't see where you would have a problem recording with it.

If you want to hear what it sounds like used for recording, just listen to any Stanley Clarke album. School Days is a good example. He runs his bass through the F2-B and into the console.

[Chadwick]

'mmmmuuulltttiiply' + viability of a wallwart [ac adapter] supply

Ya, cascade looks like an old 30s term based on a google search.

He could just use the transformer for the plates and rectify and regulate the wallwart AC votaget to run the heaters.. Keeping noisy wall voltage clear away from the unit.

With a 12 volt regulator I believe you would only need 150ma, giving more wall wart options after you 'loose' current in 'multiplying' the voltage.. [you guys are picky]


Rob, the filiment supply still doesn't look functional to me. I see that the led could go off and indicate by itself but the IN914? The scheme doesn't show where they hook up the filiment, maybe after the 68ohm resistor.

So what is the equation for current loss in a multiplying rectifier?

If we start with 18vac @ 500ma and we multiply it 7 times to achieve apprx. 126 [this would rectify upwards though I'm not sure if the 1.4 would apply, might act more like a half wave rec???] . Then divide 500/7 leaving 71ma.. wait, actually youll run the heater before the multiplier so lets take 150 ma off the top and 350/7 = 50ma for the plates which should be plenty for just two stages I believe. Rob hows the math on that? I'll bet we loose a couple more ma in the circuit but with enough left over for a 18vac 500ma wallwart to actually run a bass pre.

[RPmanen]

It does show where to hook up the heaters, just to the left of the 250 Ohm pot, pins 4,5 and 9 (CT). Or am I just loopy?

Rob

[Chadwick]

ya, your right

ya, that makes a lot more sense.. those diodes were bugging me but if you hook it up just where it says its all good.

not loopy, right on the money.

[David Schwab]

By the way, Alembic has some info on the F-2B:

Circuit detail

The F-2B was inspired by the Fender Dual Showman amplifier. In the late '60's, we routinely added preamp output jacks to Dual Showman amplifiers and substituted an external power amplifier for the built-in quad 6L6's, most usually a MacIntosh 75.

The F-2B came about to provide the same preamp functionality in a rack-mounting package, as the Dual Showman was considerably wider than the standard 19 inch rack.

We took the opportunity to make a small improvement in the input jack circuit, while providing identical functionality. Plugging into Jack 1 alone gives full sensitivity, while Jack 2 alone is attenuated 6 dB. If signals are plugged into both jacks simultaneously, mixing resistors combine the signals and give isolation to the two input sources. In the original circuit, the mixing resistors are always in series with the grid of the first stage, and contribute a small amount of Johnson (thermal) noise. Our jacks have additional contacts which completely switch out the mixing resistors when using Jack 1 and the ultimate quietness of the tube is preserved.

As in its predecessor, the instrument signal is amplified by the first stage before any volume control. Without the wide dynamic range provided by the 300 volt supply, the first stage could be easily overloaded by large transients which are characteristic of electric instruments.
...

Seems like the schematic is at the very least missing the other input jack. And what kind of power transformer did it use?

Here's some pictures of the insides:

http://www.alembic.com/club/messages...tml?1076361570

[R.G.]

Hi there,
I've found the Alembic F2-B schem at Geofex, and would love to build one. The only thing is, I can remember reading about a problem with this schematic.
Rob

It amazes me a little how full of gossip the internet is. Several people have complained that I didn't specify a power transformer for that schematic. I couldn't - the original was a custom unit, made for the box. And there is almost never a mention of that schemo without someone vaguely remembering something somewhere was wrong with it.

It does match the box I traced, if that's any help.

If I was building that, I'd get two 120:120:6.3:6.3 toroids at Digikey, and power the mess from one of them with paralleled 120V primaries (I'm in the USA), run the filaments from the center of the series secondaries, and hook up the 6V windings of the second toroid to the 6V windings of the first one to power the two 120V primaries of the second toroid in series. This gives something less than 240Vac, which can be full wave rectified to around 300Vdc.

You could go nuts with DC filaments, elevated filaments, etc, but just running the 12AX7s from 12Vdc centertapped and grounded will be pretty good.

I'll try and post the cascade rectifier here.

Google "Cockcroft-Walton" and "multiplier".

With a 12 volt regulator I believe you would only need 150ma, giving more wall wart options after you 'loose' current in 'multiplying' the voltage.. [you guys are picky]

It turns out that you have to be picky on this one because all of the losses add up to drop that secondary voltage.

Rob, the filiment supply still doesn't look functional to me. I see that the led could go off and indicate by itself but the IN914?

Trust me - it's functional. The 1N914 is a 100V/100ma diode. LEDs conduct with forward voltage of 1.5 -2.5V but they don't withstand reverse voltage well at all. The 1N914 keeps the LED from breaking over in the reverse direction and being damaged.

The scheme doesn't show where they hook up the filiment, maybe after the 68ohm resistor.

Actually, it does. You see that 4, 5 and 9 to the left of the 250 ohm hum balancing pot? Those are the pin numbers for the heaters on a 12AX7. Notice that the schemo says that only one channel is shown. The second 12AX7 will have the same pins connected to the same places.

[Steve Conner]

OMG, Stanley Clarke! I'm about to try and build a bass preamp to go with a Class-D power amp I recently acquired, and I can't decide whether to try the Albert Kreuzer JFET preamp, or one channel of the Alembic F2B. I'm a big fan of Stanley Clarke, though I think my favourite tone is one I heard on a live recording of "All Blues" and I have no idea what he used for that. Probably mostly raw talent

If people complained about the F2B lacking output, I guess it would be because the output jack is fed straight from the tube plate in the published schematic. That gives a very high output impedance, which would be loaded down a lot by the input impedance of whatever you connect. A cathode follower stage, or the guts of an active DI box, would help. I see Alembic built the guts of a DI box into the F1X.

PS: I wouldn't care to try using a Cockroft-Walton stack to get the B+ voltage from the filament voltage, if that was what was being suggested. It might be feasible to use a C-W stack in a hybrid amp that has, say, an 80V CT transformer to power its solid-state output stage. (And it would look even more attractive in my Class-D amp with its +/-60V switching power supply, since C-Ws work a lot better at high frequencies, and there is plenty of HF voltage there to be tapped into.)

[RPmanen]

If I was building that, I'd get two 120:120:6.3:6.3 toroids at Digikey, and power the mess from one of them with paralleled 120V primaries (I'm in the USA), run the filaments from the center of the series secondaries, and hook up the 6V windings of the second toroid to the 6V windings of the first one to power the two 120V primaries of the second toroid in series. This gives something less than 240Vac, which can be full wave rectified to around 300Vdc.

Sorry, not quite following you . why not use a single 120:120:6.3:6.3 and rectify the 120, and use one 6.3 secondary per tube (when using 2 channels)?

Rob

[R.G.]

Sorry, not quite following you . why not use a single 120:120:6.3:6.3 and rectify the 120, and use one 6.3 secondary per tube (when using 2 channels)?
Rob

For safety issues primarily.

You could do what you suggest. However, you would be using two 120V windings which might be wound side by side. This is OK for normal primary use, as the voltage difference between any two points is never more than about 350V peak. However, if you then use one of those windings as a secondary, the two wiring insulations can't be counted on to absorb 1kV to 2.4kV line transients without puncturing and exposing your own personal body to the AC line. Bad juju indeed.

This may seem like quibbling, but let me tell you about my friend Ivan. He was the worlds champion Russian Roulette player. His career record was 453 - 1.

In additinon, you'd still need to double the 120Vac winding to get up in the 300V range. Not as good as full wave rectification.

In addition, you really don't need to worry about using separate filament supplies per tube. They'll work fine connected to a single 12Vac filament supply. Or to a single 6.3V supply.

[RPmanen]

Good Point

However, where I live, we've got 230VAC @ 50Hz. What would you do in that case? Since I already have 230 to begin with, I don't need douple 120 secondaries.

Rob