TNT Preamble Modification

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello,
I just built a very nice little 6V6 tube amp, but I want to try something different, something solid state, and something very simple. While building the tube amp, all this high voltage, excessive gain only to chop it down with an output transformer just seemed so... well... unnecessary. Although, it does sound quite nice, I have to say.

I came across this very cool preamp:
TNT PreAmble - DIY solid state stereo preamplifier [English]

I was wondering, can I make this into an integrated/power amp, enough to drive an 8ohm speaker by replacing the low power bipolar transistor with a power Mosfet such as IRF530PBF? Of course, I would need to beef up the power supply, so I was thinking of lifting the power supply from one of the Pass' Zen amps.

I only need about 4W on peaks, so I am trying to avoid unnecessary gain. My DAC puts out 2.8V, and I really only need 5-6V driving my speakers.

If this is viable, what would the potential cons of such a circuit be? Also, what would the important parameters in searching for the best MOSFET be? Low Ciss? I'm very much a newbie here, so your help is very much appreciated.
 
Why not build something like a low power First Watt F5 or a mini alephJ? you don't need a preamp,just a volume control. Not many more parts and the power version of the preamble wouldn't work without major mods that would be a new amp design.

You could also build something like an opamp set to a gain of 3 or 4 with a power follower capable of delivering an amp or two. Or add a bipolar or mosfet power follower to the preamp you linked to.

See the projects section of www.passdiy.com for some good learning material and the zen series offers some interesting low power options to help yo along your learning curve.
 
The issue is biasing the output fet. Not sure exactly the operating point of the jfet, but assume you choose one that self biases to 3 ma, which puts the drain at 10V. Connect your mosfet Gate there (with a 221R stopper resistor) and its Source is at 6V since Vgs of an IRF610 is around 4V. You have a maximum of 6V swing peak to peak, which in reality means you'll be lucky to swing 1-2V cleanly. Also, single ended means you must set the bias at least as high as peak output current. Say we choose 1.5A bias , so R109 must be 4R, dissipating 9W. A Source follower just works a heck of a lot better push pull or into a higher impedance load.

An example of a power buffer is the output section of http://www.diyaudio.com/forums/diyaudio-com-articles/150438-burning-amplifier-ba-2-a.html?garpg=3, ignoring the feedback (out) connection. For your purposes, +/-20V and a single pair of well heat sunk IRF610/IRF9610 biased at 50-100 mA would do, but IRFP240/9240 wouldn't be a bad idea. You can do it with bipolars if you reduce the bias voltage.
 
The issue is biasing the output fet. Not sure exactly the operating point of the jfet, but assume you choose one that self biases to 3 ma, which puts the drain at 10V. Connect your mosfet Gate there (with a 221R stopper resistor) and its Source is at 6V since Vgs of an IRF610 is around 4V. You have a maximum of 6V swing peak to peak, which in reality means you'll be lucky to swing 1-2V cleanly. Also, single ended means you must set the bias at least as high as peak output current. Say we choose 1.5A bias , so R109 must be 4R, dissipating 9W. A Source follower just works a heck of a lot better push pull or into a higher impedance load.

Thanks a lot for the feedback. This is exactly the stuff I'm trying to figure out. I need to hit the tutorials, and figure out the operating points and do some proper calculations, which I am learning to do. What you say is very helpful. Nelson's site is awesome. I've been digging around there for a while. I also found some interesting single ended designs here btw:
Project 83 - MOSFET Power Follower

I just built a PP amp, and was thinking this time to go SS single ended.
Thanks
 
Rod's got a lot of good information on his site, and some nice projects with boards available. His follower would still need some sort of preamp to deliver more than a couple volts your DAC puts out. If you build that buffer move R1 to between the connection of the 15V zener and the mosfet gate. In this position it is called a gate stopper, physically as close as possible to the gate to help prevent oscillation. Mosfets can sometimes misbehave looking directly at a zener. If you were to use the preamble to drive it, you could eliminate R0 and C1.

Micro tutorial: Note the darlington and zener below the mosfet in Rod's Project 83. That forms a current sink, which offers a higher impedance load for the mosfet than R109 in our preamble turned amp. The Source is also lifted up to 19.8V, allowing much more swing. Note that you need some hefty heat sinks for this project, with 88W dissipation.

A mini Aleph-J running 20V rails or so would get you the little bit of gain you need and keep you in single the ended realm. I believe Peter Daniel still has boards available. Can you tell I am a Pass fan? ;) You can get away with half the heat sink you would need for the project 83 buffer.
 
Last edited:
The mini Aleph-J does seem pretty cool. I was not aware that there was a mini version.

BTW, my original idea of subsituting the bipolar for a high power FET and driving a speaker directly will not work very well for another reason. The input capacitance of any of these high current FETs is pretty high at least 1500pF, so I need to drive it from a very low impedance source. I suppose thats the whole point of the bipolar in the Preamble- to do just that.
 
I suspect you are thinking of the larger parts like the IRFP240. The IRF530 Ciss is only 600 pf, and IRF510 is 180. The issue with high Ciss will be a softer high end, maybe even a roll off. Not really an issue at my age, but if you are young it might be.

Any of the Aleph boards can be a mini if you just use a pair of output devices. You may have to do a little lead insulating and rearranging to put jfets in a board designed for IRF9610s but it's not that hard. The circuit is simple enough to build on veroboard if you want, especially if you just use a 2SJ74 as your CCS.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.