Can someone explain the 3 Harvard inputs?

[rp]

I should be smart enough to figure this out, but I'm not. Is it set up Low, Med, High? And is it: #1 sees 3 68Ks in parallel for 22.6K, #2 sees 2 68Ks in parallel for 34K, and #3 sees 68K?

And, why no 1 Meg load resistor? Were they trying to keep the gain down to keep it clean?

I'd be temped to just go with the standard 2x68K+1 Meg, but I played through a Harvard long ago for a few minutes and remember being impressed. There are YouTubes of Harvards and Victorias sounding great too, so there must be something to the stock set up.

[martin manning]

Like this, I believe:

Input 1 (High)
Series R 22.7k
Shunt R Infinite
Attenuation 1 (0 dB)
Impedance Several Meg?

Input 2 (Med)
Series R 34k
Shunt R 68k
Attenuation 0.67 (-3.5 dB)
Impedance 68k

Input 3 (Low)
Series R 68k
Shunt R 34k
Attenuation 0.33 (-9.5 dB)
Impedance 34k

With the standard 2x68k and 1M, #1 gives you 0 dB attenuation and 1M impedance, and #2 gives you -6 dB and 68k impedance.

[tonewood]

Looks like the 68k's aren't just different grid stoppers but they act as voltage dividers also.

input 1: no divider and a stopper of 3 68k's in parallel.
input 2: divider with 1/3 attenuation
input 3: divider with 2/3 attenuation

I'd also would like to hear about why no 1M ref resistor.

edit: M M said it more clearly while I was trying to figure it out.

[martin manning]

A 1-Meg resistor could be added to input #1 to keep the ground reference close when that input is being used, and I don't think it would have any effect other than helping to stabilize the bias point. As it is, it relies on the path to ground through the guitar.

The series resistance and the Miller capacitance of the tube determine the knee of a HF roll-off, so the three inputs will differ in that regard, but this effect will be subtle. Assuming an input capacitance of 100pF puts the -3dB frequencies outside the audible range.

Are you using a 6AT6 or some other tube for the input stage?

[rp]

Are you using a 6AT6 or some other tube for the input stage?

I'm eBaying some 6AT6s and 6AV6s as I type. I would never think of modding to a 9 pin, I love using different tubes and that was part of the attraction to the 5F10, feebly unique though it is. Wish I really knew what I was doing, I'd do the Shindo thing and build with all sorts of abandoned misfit tubes.

So, if I understand this right, regarding the load resistances and gain, this triple input isn't really very different than the standard 2 input?

Great answers, BTW thanks. Fender engineers knew their stuff as does Martin!

[martin manning]

So, if I understand this right, regarding the load resistances and gain, this triple input isn't really very different than the standard 2 input?

No big difference that I can see. The 6AT6 tube may well be part of the signature tone of this amp, so if they are available go for it. Yea, how many of the electronic devices manufactured in 2010 will still be around in 2070?

[Bear]

You've got a real one? Mazel tov.

Otherwise, I would point towards the tweed Vibrolux circuits. Essentially the same amp as a Harvard except using 1/2 a 12AX7 instead of the 6AT6. Plus bias-vary trem. Even if you don't build with or use the trem, a hell of a convenience factor. But you lose the coolness of bucking modern sourcing restrictions.

Mmm. Green onions . . .

[dobbhill]

I have a 5F11 Vibrolux. I added a 1meg resistor to the input and found that it provided a more appropriate load for the guitar, improving tone markedly, especially when using the volume control on the guitar.
Hope this helps...
D

[rp]

I have a 5F11 Vibrolux. I added a 1meg resistor to the input and found that it provided a more appropriate load for the guitar, improving tone markedly, especially when using the volume control on the guitar.
Hope this helps...
D

That helps a lot, now I'll probably try it, it's certainly not a lot of effort.

I really want to build it with just 2 inputs, I'm sure the choice of 3 was the Sales or Marketing dept looking for something to help the salesman explain the Harvard next to a Princeton and Deluxe - "Well sir, this one's got three inputs!" But for all I know Cropper plugged into the odd one (#3?) so I'll build it straight. Plus it will be a 'coolness' test, it's going into a ss Marshall 10" carcass, so anyone in the know will be tipped off by the 3 inputs that it's a Harvard.

[mcrracer]

Does a separate 1M resistor need to be added to each input jack?

[Cliff Schecht]

If there was an input capacitor then yes, you would need the 1M pulldown resistor. Because it's direct coupled, they are essentially using the grid-source resistance as their resistor, although in practice this is a terrible idea. If any of your effects circuits have a DC blocking cap and no pulldown resistor then you'll slowly saturate the input as the capacitor charges up over time (literally looks like the sine wave shifting up until the entire thing clips). The input resistance is in the TOhm range and looks like an open to the cap, allowing it to charge as if no load was attached at all.

[martin manning]

Does a separate 1M resistor need to be added to each input jack?

No, only on input 1. just like on the usual 2-input high-low arrangement the lower leg of the divider formed when the attenuated inputs are used gives the grid a ground reference.

[mcrracer]

Thanks MM, Just wanted to be sure.

[Richie]

most old amps had the inputs labeled Mic, Accordian, Guitar. May not be in that order but they were ment to use different things.

[M Fowler]

Thats because we put the amp in front of us in those days and plugged in our guitar and mic to the amp.

On some gigs there was only one amp shared by bass, guitar and mic. No accordion in my bands, not that I didn't like them.

The first PA with monitor if you will.