The Voltage at the minus side of C1 should be half the 9V battery Voltage. If the Voltage there is too low, increase the resistance of R2. If the Voltage there is too high, decrease the resistance of R2.

If Q1 and Q2 are too hot, install a resistor across one of the diodes, D1 or D2. Try a 1K or 470 Ohm or go lower until Q1 and Q2 cool off.

This circuit is too simple. It won't work at all 99% of the time. If it does work, performance will not be very good. An LM386 is easier to use and will work most of the time without adjusting any resistors.

The preamp is another crappy circuit. Find something else.

Big drills typically have limited speed (RPM). For small diameter drills you need higher speed.

The tubes that IC's come in (if you buy enough) generally do not provide ESD protection for the parts, they merely don't generate static. The shielding bags that are generally dark metallic gray or aluminum colored do provide protection. The pink bags and pink bubble wrap generally will discharge any static buildup, but aren't as good as the shielding bags at preventing all damage.

The problem I've had was when the humidity got too low. That's when you start drawing sparks when reaching for doorknobs. MOSFETs are the most sensitive, especially the smaller ones. I use anti-static mats at all work areas. I use a humidifier when things get really bad.

The most basic piece of test equipment is a digital Volt meter (DVM). You can usually find cheap ones for around $10 USD. Sometimes at Harbor Freight they give them away if you buy something else, even a pack of 9V batteries.

Data I have from the book Essential Characteristics:

12B4A   Low gain preamp triode can work with 6.3V heater
12CA5   Small power tube good for 1.5W
12AL5    Dual low current diode
35C5     Small power tube good for 1.5W  Needs 35V heater supply
12BD6   RF Pentode might work as a preamp
12X4     Rectifier tube
12FX5   ? Did you mean 12FX8?
12BDK5  Did you mean 12BD5?
12BA7   Radio tube not useful for guitar amp

When the chips go bad, you can usually find an internal short by Ohming between the V+ or V- pins to the output pin. Measure an unused chip to establish what a normal reading is. On a good chip the reading may change if you reverse the red and black meter leads.

The tab of the TDA2050 is connected to the negative supply pin so the package must be insulated from the heatsink.

An interesting Nelson Pass design called the PLH is worth a close look. It starts with the JLH architecture but is simplified to reduce the amount of feedback and uses MOSFETs. There is a pot that lets you adjust the amount of low order distortion. You can find the article he wrote on one of his websites.

Link: https://www.passdiy.com/project   scroll down to PLH and click the More button.

The PLH uses a 40V rail, so power dissipation is pretty high requiring a huge heat sink and parallel output devices. It will work on lower Voltage, 24V to 30V or even 19V will reduce heat and output power.

The ACA that printer2 posted above is nice because it uses a common 19V laptop power supply. That makes it easy for the novice to get running. In the schematic posted the feedback is eliminated and the amp produces sweet sounding even order distortion.

7815/7915 regulators are not as bullet proof as the manufacturer would like you to believe. Short +15 to -15, one regulator wins and the other dies. Think they will just shut off if they get too hot? The one that shuts off first will likely be killed by the other one. Just use zeners.

Agree with Enzo, check those screen resistors even if they are new.

Hey, you gotta like that he has two scopes!

The same circuit was looked at in this thread:

https://www.ssguitar.com/index.php?topic=3373.0

Quote from: g1 on January 26, 2016, 11:59:27 AM
  To speculate a bit about the etymology of "rail":
in electric railway (and subway) systems, the power is taken from the "third rail".  This goes back over 100 years.

Good point! Same with Lionel toy trains where I got my first electric shock. Really just a tingle from 24VAC.

There was a discussion on MEF a year or so back about a "Split Rail" tube power amp. In that design the screen supply is about half the main B+ Voltage like many of the Music Man amps.

Many semiconductor manufacturers are discontinuing products that have low sales numbers. Frequently it's the through hole version that gets the axe first, OEMs only use the surface mount versions and that's where the big sales numbers are. Some of the new parts that get designed are never made in through hole versions. There are little adapter boards that you can solder surface mount IC's to for prototyping.

Is the Buz 901 part a lateral or vertical FET? The difference being the Source is connected to the case on a lateral (2SK134) and the Drain connected to the case on a vertical. Biasing is completely different too. Unless you can find some laterals, circuit mods will be required.