Hannes123
Active Member
Problem Solver
Continuing on, the only practical way to really solve the level control roll off problem then is to install a SS buffer circuit between the level control, and the input to the power amps. Since the level control has a loudness circuit associated with it, that makes it a part of the preamp section -- as opposed to being part of the power amp section. Therefore, the buffer circuit should be fed by the level control, with the output of the buffer feeding the new Preamp Out Jack. With the jack straps in place, that means that the consistently low output impedance of the buffer (regardless of level control setting) will then directly drive the input to the internal power amps, and remove the concern for Miller. It also means that when the straps are removed, the Preamp Out Jack represents a nice low impedance source to drive any other power amplifier you might wish -- solid state or otherwise! That lets the 400 operate as a very nice tuner preamp control unit for great versatility in any system.
As for the buffer circuit itself, an op-amp set up to operate as a unity gain amplifier offers the greatest performance return in this application. With good power supply noise rejection, and the capability of very low output impedance, it's a package that's hard to beat. A bifet device was chosen because of its extremely high input impedance, with that figure being in the trillions of ohms. Also, since the chosen device is a dual inline package (DIP), it's very small, and has two op amps built into it, which is perfect for the application. Best of all, being that the device is a TL082/LF353N, you can run right down to the nearest RS and pick one up since it's a store inventory item.
For those concerned about sonics with an op-amp in the signal path, you shouldn't be. Being that the buffer circuit offers no gain (employing 100% NFB), it's about as neutral sounding as any circuit can be, and solves bucket loads of problems, while introducing virtually none. Therefore, it is quite responsible for actually improving sonics, rather than hurting them.
With a basic plan developed then, a small R/S perfboard was used to build the buffer circuitry, and space was made available behind and to the right of the balance control for mounting. This was accomplished by relocating the two 25 ohm 'Speakers Off' load resistors to the other side of the terminal board they connect to, leaving the space between the balance and level controls rather open. The small perfboard could then be mounted in this area with a small L-bracket through one of the cooling holes around the tuner can cap, meaning that once again, no physical alterations were necessary to install the circuit. The location was also ideal, since it allowed for only 3.5 inch shielded leads to connect the signal between the level control, and the input to the buffer amp board. Shielded cable can also cause excessive HF roll off in high impedance circuits if they have any length to them at all, and keeping them short in this case is icing on the cake. With the plan off attack for this problem nearly finished then, all that was left was to figure out how to power it. Remember that neatly folded up white wire that used to supply bias to the output tubes before EFB was installed?
It was perfect. It is relatively well filtered -- well enough for an op amp anyway, and was just lying there, supplying about -18 volts to.......nowhere. Tying the buffer circuit to this source dropped the available voltage down to about -15.5 vdc, which will allow it to develop 5.0 vac RMS at the Preamp Output Jacks if necessary, which isn't necessary for the internal power amps, since they are driven to full power output with just under 2.0 vac RMS at the Power Amp In Jacks. Also, a 5 volt RMS output level will certainly drive almost any external power amplifier to full power output as well if the 400 is used as tuner preamp unit only. Altogether then, the buffer circuit became the perfect answer to the problems imposed by Mr. Miller, and boy oh boy did it solve the problem.
Pics include:
1. The basic buffer circuit board before connecting leads were attached.
2. Mounted and connected up in the space created as discussed.
3. What a difference a buffer makes! Here is the same 10 kHz square wave produced with the level control in the same position as before, driven by the tone stage amplifier, but this time with the buffer installed and operating. It does not change its shape with rotation of the control -- only its amplitude. There simply is no comparison.
4. The schematic of the buffer amplifier, with one channel shown, and the terminals within the DIP representing the other channel.
To be honest, this part of my 400 project was quite challenging -- not only in developing a practical and workable solution that really solved the problem, but also in building and installing it so that the finished job was neat and professional.
The effort was well worth it however, as removing this compromise takes the 400 completely out of the 'receiver' performance category, and moves it to a level where few other pieces of equipment reside. Just as a chain is only as strong as its weakest link, improving all the individual elements in the power amplifier section proper won't solve the problems created by the original level control design. Installing the buffer circuit does however, and then allows the power amplifier modifications to really shine, delivering a realism I have never heard out of a Fisher product before.
Obviously, for those that implement this modification, it does require skill to work with a small circuit board -- or with whatever format you might chose to build the circuit on, and installing it is somewhat tedious, but again, well worth the effort. The basic circuit and idea should be very adaptable to many Fisher products to enhance their performance as well.
In resolving this issue however, one last opportunity to wring out maximum performance from the 400 presented itself -- very much in keeping with the proverbial thread I've been pulling on that I mentioned at the beginning of the EFB thread. By comparison, it is rather simple as compared to the buffer modification, but again, makes the 400 even better, more practical, and much more up to date as well. Next time. But thank heavens, we really are about out of thread!
Dave
EDIT: The Buffer Amplifier circuit has now been revised, which is presented and discussed in post #345 of this thread. The revised design will only offer improved performance in selected scenarios as discussed in the text.
so cool and instructive - asking for a friend, would this also make sense in a Fisher x-100 (early model) ?
