QUADII with DHT

[revintage]

Having two old QuadII, that I personally do not enjoy listening to, and a good supply of NOS 6B4G lying around I am just about to start rebuilding.

I have already taken the OPTs out of their cans by cooking them for a few hours at 150 degrees C to melt the goo.

As the 6B4G can not be loaded with 3350ohms in PP I will rewire the secondaries adding some extra soldering tags. This makes it possible to try 7600:8ohm instead. Wired for the original 8ohm connection will load the primary with 6700ohms when used with a 16ohm load.

If the first try works out like I hope, the next step is to change the EF86s for 5842 or 6C45. After that I will also try rewiring the primary and omit the cathode feedback windings. This will load the primary with 6100ohms with 8ohm load instead. With 16 ohm load on the original 8ohm connection the load will be 5500ohm.

As the amp will be used for midrange horns the lows are not important.

[revintage]

This is the simplified version with the cathodewinding omitted:

[JLH]

I would not be so quick to remove the cathode feedback. If I recall correctly, the each cathode feedback coil is about 15%. If this is the case, then by using them it will effectively drop the 6B4's Mu to 2.5, and its plate impedance to 480 ohms. This way you would not have to mess around with using different secondary taps to get the desired reflected primary impedance.

Rgs, JLH

[EC8010]

Don't forget that if you use cathode feedback with DHT you will need separate heater windings for each DHT.

[revintage]

Thank you guys for your comments!

EC8010: Didn´t you read the first schematic? What do think 6VDC 1 and 6VDC 2 means ?

JLH: If you also read my text carefully the first try will be with the cathode feedback . My concern about not using it, is that the demands on the driver gets very high with the windings in place depending on, as you say, that mu is lowered.

The anode-winding is 2880 turns and the cathode-winding is 320 turns. This would mean the mu and Rp isn´t lowered that much, I guess?

Brgds
Lars

[EC8010]

Whoop, didn't spot that - it looked like the standard diagram.

[andyjevans]

Consider using DHTs throughout. For instance the 3a5 in the first two positions. You can then build a balanced amp all through, and for example direct couple stage one to stage 2, then transformer couple to the outputs with Lundahl LL1660 or similar. Again, you will need seperate filament supplies for each 3a5. You would need to consider a second chassis for the filament supplies.

All DHT amps are pretty awesome in terms of detail and delicate treble.

[revintage]

Andyjevans: Thanks for your concern! The whole idea is to rebuild the Quad without adding any extra boxes. Also I want to keep as much as possible the original topology with only two gain stages. So an IT(though it crossed my mind ) is out of the question, as is the 3A5 that has to low mu, lousy linearity and low Uamax for this application.

Brgds

[revintage]

JLH: After reading your post again, I realize that I don´t fully understand what happens impedance-wise with the cathodewindings in place. Could you please help to clarify. What will the reflected primary impedance be?

Any how I did a spice sim of the circuit and the EF86s are in trouble ! It is hard to get enough headroom.

[revintage]

I also have a pair of Lundahl LL1676. Think they might work well at the input wired 2:2+2. This would make the amp completely balanced. They could be fitted on the front where now the 6pole jack is.