Quote1. Regarding the power supply: My intention is to hook this up to a 8Ω cabinet. So, are you saying that the original wall-wart will not be enough? If this is correct then, what are the specs for the proper PSU?

I think earlier you had mentioned increasing the power. You don't have to, but it can handle a larger power supply if you want. The max the output op amp can handle is 20W and 44Vdc. For an 8 ohm load, you should be able to go up to 44V. However, my earlier warning was if you are going to try a 4 ohm load, then stick with 34Vdc because obvioulsy the current will be larger for a 4 ohm load than for an 8 ohm load. Increasing the power supply to 20W won't make a huge difference but it will be slightly louder at maximum clean output.

Quote2. I thought phatt's advice about removing the diodes meant the gain pot is still working and you still get some distortion, you just get less distortion or "cleaner" distortion. However, from your last post, if I understand right, are you saying the gain pot becomes inactive and you basically lose ALL distortion this thing can produce? If that's the case, then no this is not what I am after. I do want the versatility of having a Gain control before my Volume. Question is can this Gain control be more gradual in response and smoother in performance? 

Essentially yes, but it is a bit more complex than what I explained. The diodes clip the output of the first stage and cause harmonic distortion. If you remove them you can still get distortion from overdriving the amp. But the purpose of the first stage is merely to overdrive the clipping diodes when the gain is set high. In my opinion the best solution is to simply reduce the maximum gain of the first stage so that the gain pot becomes more usable. However, you can try both solutions and see which one suits your needs.

Quote3. The 24V 600mA wall-wart powers your MA007 right? Are you using it as is or you modified it and play it through a different speaker?

Is this the MA007 ? Is it made of plastic? I'd be curious how it sounds!?!

Yes. I did not mod the new wall-wart. My original intention was to mod the original wall-wart with a different transformer but it turned out to be simpler to just replace the whole wall-wart. Yes I have the MA007. I actually have two. One that I moded and I bought another to compare the original sound. They are different. But I don't have the ear to really say which is better. The moded one is slightly more powerfull with clean sound and my volume and gain controls are more usable. I also moded the tone stack to give it a brighter sound. And I put a large filter cap to reduce the humm.

Your last statement is not correct. You can hook up a 4 ohm or an 8 ohm speaker to the 2030A.
You can replace the wall-wart to a 20W ( more would be a waste since the 2030A cannot handle it ).
The problem is to figure out how high of a voltage to get for the wall-wart.
There are two problems:
1 - the rating on the wall-wart is nominal. It may be more than the rating even at full load.
2 - the current drain of the 2030A is dependent on the supply voltage, the load and the input signal. And hence, so is the power dissipation.

The 2030A is rated for 44V max and 20W max. However, the current is dependent on the load. If you use a 4 ohm speaker then a 34V supply will dissipate 20W. I=P/V so the current is about 600mA.

What I can tell you is that I have replaced the wall-wart with a 24V 600mA and it works fine.

If you want to create a cleaner sound with the gain at 0, then you need to decrease the gain in the first stage of the amp. The designers set it too high. The clipping diodes that you were told to remove begin distorting the signal at about 800mVpp. You could just remove these diodes as suggested ( d1 and d2 ) if you don`t want to be able to use the distortion circuit. Another solution is to increase the negative feedback to the stage 1 amp. This can be achieved by replacing the feedback resistor with a higher value. The feedback resistor is in series with an electrolytic capacitor between either pin 2 or 6 ( depending on which one they are using as stage 1 ) of the JRC4558D and ground.

Yes, you're right. I was more concerned about the 20W power dissipation but looking at the test circuit output power it should be able to handle more. It depends on the load though. At 4 ohms it reaches 20W at 34V. For 8 ohms it is above 40V. Don't forget that these are actual power supply voltages not the nominal values rated on the power packs.

The 1.5A power supply is stiffer and so is more resistant to power supply humm. You could also reduce the humm by replacing the filter cap on the PCB with a larger value.

The 14V power supply is a nominal 14 Volts. It may be 20 or more V when unloaded but it is much less when fully loaded at 500mA. You can increase the wattage of the amp slightly by replacing the power supply with a higher voltage power supply. I am using a 24Vdc at 600mA on my MA007. The 2030A cannot handle more than 20W so I don't suggest going higher than 24Vdc.

If you want to increase the range of the volume control you can replace the resistor from the -ve pin of the 1st stage of the JRC4558 amp to ground ( through a series cap ) with a higher value. On mine it is reference designator R3 but it may be different on other models. On mine its a 270 ohm. Try replacing it with a 1 kohm. This will decrease the full scale clean gain from 400 to 100.

It's also a good idea to buy some heat sink compound and re-apply to the new IC. Clean the old guck off the heat sink first. This will help prevent the new IC from burning out quickly.

Is there any 'real' evidence that non-polar caps are better than electrolytics in signal path designs?

 

In power supply design, it is important to calculate the surge current when the filter cap is larger than 1000 uF. A common design is to insert a series surge resistor to limit the surge current rather than using more expensive and larger diodes.

If your TD3050 had a bad solder joint, that could have caused arcing and produced the surge current that destroyed the diode. The shorted diode, in turn, shorted the secondary of the transformer, which blew the fuse. 

Attached is the schematic for the MA007, almost identical to the MA004 but with a few upgrades. It uses the TDA2030A instead of the 2003 and some changes in component values. Note that the reference designators are almost all different.

A few comments. It is common that op amps do not go rail to rail. This is especially true when the load is heavy. The term is output compliance and it is dependent on the load. Also, diodes are only linear at small signals. The 1N4148 for example, begins to show distortion on the scope at around 800 mVpp, much below the clipping voltage of 1200-1400 mVpp.