Ampeg inductor substitute, V2 build, various questions

[JazzGuitarGimp]

I think the first thing I'd try is adding a series resistor on each side, between the 0.33uF caps and the 100K grid leak resistors. Making them 100K will cut your p-p voltage drive in half.

[Smokebreak]

Thanks Lou, that makes good sense. Make a voltage divider with the grid leaks.
I got the heaters wired up this morning. I'm using the large 9 pin sockets from the organ but hedged my bets and used new ones for the preamp.
So far with the chassis and sockets I'm at $50. Gonna try and make it work for under $100.

[JMFahey]

So, since I'm using the 7868(7591) tubes in this V2 build, and the V2s ran various 6L6 types, what would be a good approach to reduce the signal to the 7868s, as they are a higher sensitivity tube than a 6L6 type? Clearer scheme courtesy of jaysg

Don't overthink it.
1) the difference in sensitivity amounts to much less than 1 dB to begin with.
2) and said tubes are inside of a negative feedback loop, which reduces such difference to tenths of a dB , absolutely inaudible.
3) any messing with the circuit, such as adding (unneeded) resistive attenuators, will change the load seen by the PI and cause a far larger, unpredictable effect, so best choice is leave as is.

Just as an example, an EL34 is far easier to drive than a 6L6 , and yet nobody compensates when going from one to the other (for good reasons).

The only one who does such a thing is anal tube designer Mesa Boogie in their simulclass amps, and they are forced to do that so widely different (I'd straight call them incompatible) tubes "push the cart roughly the same".

[JazzGuitarGimp]

I had assumed we were talking about a substantial difference in drive requirement: i.e. -42V bias for 6L6 and -14V bias for 7868 - in which case the difference in drive voltage is about 9.5dB. That's what I get for not looking at the datasheets.

[JMFahey]

Ouch!!, same here
I assumed they were roughly similar.

Something else I didn't mention is that adding a 100k resistor in series with 100k grid reference ones will combine into a 50k effective grid stopper.

I bet *that* will change amp response in a very audible way, nuking high frequency response.

Of course again I didn't look at the datasheets, so 7868 in fact might have 5X smaller Miller capacitance, in which case I'd be wrong again

That's what happens when people replaces usual tubes with much less common ones.

Oh well

[JazzGuitarGimp]

A 2:1 divider will get you 6dB of attenuation; 4:1 will get you 12dB of attenuation. Once the needed amount of attenuation is determined, the two resistor values can be scaled up for a net 100K grid leak (or 220K, or, whatever value is needed to overcome the effects of miller capacitance).

[AL]

The early Ampeg SB12's used 7868 power tubes. You may want to check there and see if it helps. Fliptops does sell the power and output transformers for both the SB12 and V-series.

I've checked before for inductors - I recall finding them (or close enough) at Mouser. McMaster Carr carries the vibration mount replacements if you need them or want to mount the amp like the old v-series.

Really interested to see/hear how this turns out. I've been running a VT40 as my main amp for several years. The midrange control is magic.

AL

[Smokebreak]

Thanks AL, I got a bunch of different inductors from Mouser, and ended up putting 8 100mH in series, and tapping the string at 100mH, 300mH, and 800mH.

Juan and Lou, stop me at any time if I'm incorrect, as I try to understand the drive voltage differences, as well as my understanding of grid bias voltage as it relates to P to P AC signal voltage..

Looking at the datasheets for 6L6GC and 7868, at typical AB1 conditions @ 450V plate, 400V screens, The 7868 wants to see -21VDC bias, which puts the P to P AC signal voltage at 42V.
The 6L6 wants -37VDC, but it states the P to P is 70V.
Now, why isn't the P to P double, at 74V for the 6L6GC(as it is for the 7868)? Doesn't the DC bias dictate the AC signal swing(as far as "typical" operation)? This is where I'm most confused. I'm having a hard time understanding the relationship between the two. I mean, I understand how the AC voltage swing affects the grid bias, resulting in amplification, but I seem to be missing something here..

Furthermore, this means that the 7868 needs about 40% less signal voltage than a 6L6, correct? So I'd be looking at a 1 : 2.5 divider with the 100K bias resistors?

So to sub 7868 in this 6L6 power section, maybe a 40K/100K divider before the 47K grid stoppers?

[JazzGuitarGimp]

Are you planning on using the V2 phase inverter circuit? If so, keep in mind that the 7027A outputs were biased at -65V (according to the V2 schematic I found at the Watts site).Trying to shoehorn that into your output tubes may require a bit of attenuation. But if the PI you're using was designed around 6L6GC outputs, I think you're close enough assuming you want the output stage to clip before the PI stage. I'd say try it and see what you get. You can always add a few more resistors if you really need them. And possibly the best way to determine the value of the resistors is to install a dual-gang pot, wired as a psuedo-PPIMV and play it for a few weeks, trying different settings on the pot. Once you've zeroed in on the setting you like, disconnect the pot and measure it to see what fixed resistors to install.

By the way, it was nice being able to log-in to TAG this morning. Thanks to admin and their crew for getting the hack sorted out. I'd hate to lose such a valuable resource.

[Smokebreak]

Oh man I've been outta town, so I've just been mostly reading, but I was wondering why the board was exceptionally slow!

Well, the Ampeg has nearly 600V plates, which I imagine accounts for the massive negative voltage at the grid, but I thought 7027A was more like a rugged 6L6GC? I don't have any experience with those tubes though. I'm expecting 420-440V on my 7868s. I like the dual gang idea to dial it in! But yep, I'm using the stock V2 PI/driver.

[JazzGuitarGimp]

Ah! I missed the fact that you're running your plates nearly 200V lower. Yes, the 600V plate supply will require a more negative bias than 440V.

The 7027A had a 35W max plate and 4.5W max screen dissipation vs. the 6L6GC's 30W and 5W ratings.

[Smokebreak]

Still got a fair amount of wiring to do, but here's the chassis as it stands. I was able to use about 50% of the board's components from the organ amp. Lotsa weird values in the V2!

[Smokebreak]

Just realized I forgot to wrap this one up. Hopefully it may help future V2 builders and 7868 users! To recap, I used parts from a Wurlitzer organ that was donated to me, and tried to use everything I possibly could from the organ, including the weirdo 9 pin power sockets. Total cost was around $100, which was basically the Watts chassis, the board Doug H. did for me, and probably $30 in random components.

The discreet inductors worked great for the mids. I used these : http://www.mouser.com/ProductDetail/Bou ... zO7g%3D%3D
..and seriesed them up and tapped out at the appropriate spots. I mounted them on this mini-turret board I accidentally bought on Ebay years ago...same with the bias board. Glad those finally got put to work!

The 7868s have plenty of life in them, and the 5BC3 is still doing it's thing. 450V screens don't seem to be a bother. I didn't think I'd need the bright switch, but It's definitely a worthwhile part of the amp, and opens up whole new doors with the mid controls in certain positions. I'll put that on a pot switch when I get a sec.

That lil Wurlitzer OT sounds good!

33K before the 100K grid leaks seems to drive the 7868s to a place that sounds good. 47K stoppers on the grids.

Amp is clean clean clean, and definitely has that Ampeg vibe, which is totally unique. I'm diggin the Stones tones. Clean, but loud, and overbearing

I just got a new phone with a good camera so I'll dump some pictures. Probably too many!

[Smokebreak]

pics

[Smokebreak]

more