12DW7 as CF and/or FX Loop

[PharmRock]

Hello all,
Finally getting back to the soldering iron....I have a couple of 5879-based preamp circuits I am contemplating at the moment.  One will probably be the DMARs with FX (in a combo), the other a single-ended (6L6) head based on either the mini-TOS or the 5879 SE circuits attached to this thread.

My question is in regards to a 12DW7 tube, which, as I understand it, is a dual triode with a 12AU7 and a 12AX7.  I've seen this tube used in a couple of different applications...one (such as the 5879 SE) where the 12AX7 drives the 5879 and the 12AU7 drives the tone stack, and others where the 12AU7 drives the FX send and the 12AX7 handles the FX return.

Is there much of an advantage of using the 12AU7 side of this tube as a CF for driving the TS or as an FX send, vs a typical 12A_7 variant, like a 12AY7 or 12AX7?  I seem to recall that a 12AU7 can provide more current in this situations. 

The direction I am leaning towards is inserting a tube effects loop after the treble wiper on the 5879 CF SE Schematic below.

JJ 12DW7 tubes can be found pretty cheap...around $15-20.  Is there much to gain by using this tube or just go with what I already have on hand (5751, 12AY7, 12AX7).

Thanks!

[pdf64]

There's no benefit in providing a greater current sourcing capability than the load can draw.

[tubeswell]

I seem to recall that a 12AU7 can provide more current in this situations. 

Depends on the load. Not much added value if you pair the AU7 with a 50k to 100k load (i.e., compared to an AX7). However if you use (say) a 10k load resistor and bias it accordingly, the AU7 will guarantee good impedance bridging in most situations - albeit that gain will be quite modest (and nil if you’re using if for CF).

[tubenit]

Here is another way you could approach the Tweed Overdrive Special mini.  NOT saying you should do it this way, just making note of this as another option. Just offering this as a consideration for you OR someone else on the forum that may have interest?

I think this would have more gain to it?  One might want to try a 12AY7 in V1 if there is too much gain?

with respect, Tubenit

[PharmRock]

Thanks Tubenit!  This is great!    I have a few extra MOSFETs from a previous build with an EF86-based preamp.

As for the loop location in the circuit...could it also go after the Overdrive toggle switch and before the power amp?  That way, I could have effects on either the clean or overdrive settings? 
I'll work on a schematic later today and post. 

EDIT...I see you already did that in the V2 schem.  Thanks!

Thanks again,
Bruce

[PharmRock]

When using a MOSFET in a CF situation, and/or for driving the FX loop as shown in Tubenit's schematics, does the MOSFET need a heat sink (chassis or otherwise)?  Or can it just "float" above the circuit board?

When I used VVRs (Hall Amplification) a long time ago, it was required to bolt to the chassis, but I understand that a lot more heat was being dissipated in that situation.

Thanks!

[tubeswell]

When using a MOSFET in a CF situation, and/or for driving the FX loop as shown in Tubenit's schematics, does the MOSFET need a heat sink (chassis or otherwise)?  Or can it just "float" above the circuit board?

When I used VVRs (Hall Amplification) a long time ago, it was required to bolt to the chassis, but I understand that a lot more heat was being dissipated in that situation.

Thanks!

Depends on the MOSFET. A bog standard IRF820 N-channel mosfet in a TO220 package has a 500V 1W rating (without a heatsink), so won't need heat sinking if used in a Source Follower circuit with a (say) 100k load resistor (because assuming a source output voltage that's about half of the HT voltage, you'll see about 1/4th of a Watt * - and if you want it to dissipate even less power, use a 220k or 470k load resistor, or a lower HT voltage)

*Assuming an HT of ~400V.

[tubenit]

Two of my amps have mosfet CF's.  And I have not used a heat sink on them and they've done fine over the years without one.


With respect, Tubenit

[PharmRock]

Thanks guys.
If I'm reading the schematic correctly, the FX loop is parallel.  I've only had experience with serial FX loops.  Tubenit...is there a particular reason you went with parallel in this amp?  Also, does the 250K pot after the FX return triode essentially serve as a master volume?

[tubenit]

I went with parallel after trying both series and parallel. So, simply a preference thing. Typically I've used a 220k to 470k resistor in the parallel approach.  One could easily use a spst to switch between those. 

Series FX sounds great also.


Yes, the FX return pot does function as a master volume of sorts.


With respect, Tubenit