SC480 Amp Questions

[Ian Finch]

Hi briansume.
I hope you mean that B-E voltage is actually ~0.65V, not ~6.5V. If so, its is OK for the bigger transistors to read lower, even 0.58V, with the small current flowing.

When the bias potentiometer is adjusted correctly for optimum bias, you should have about 2.4V DC output from the bias generator, as read across the MJE340/350 bases. It should also be adjustable, just as we discussed the bias current in the output transistors which it controls. Note that it is the bias current that is the most important.

20 mA total current in the amplifier is too small. Here is what most simple class AB amplifiers like this one should read:

Input stage 1.5 - 2.5 mA
VAS 8 -20 mA (this is a "balanced" VAS, so more current)
Output transistors 25 mA/pair for lowest THD or "optimum"

Adding these currents, a figure of 40-45 mA would be closer to correct for total current with no load, so I think you still have bias problems. Check rhe current flow in the VAS by measuring the voltage drop across the common resitor at the bottom of VAS schematic, where it connects to the negative rail. Sure, you will have audio and never have thermal problems without enough bias but the sound will be awful. ,

[briansune]

Hey Ian,could you send the upgrade MJ15003 to my e-mail address, case i would like to increase the power of the amp? graph or circuit diagram might be the best.THZ the amp are well operate. THZ for teaching me.

[Ian Finch]

I don't think you need any schematic of modifications to upgrade the output transistors. Other components mostly remain as before, so MJ15003 & 4 are convenient "drop-in" replacements for the 2N3055 & MJ2955 pair.

If you wish to use the higher rail voltages, I suggest replacing the VAS transistors with >100V rated Vceo types, as already described. There are several old colour TV video driver types which have the speed and gain required here - and you should slightly increase (~15%) the resistor in the input stage current reference to keep the small current there the same and also the common collector resistor connecting the VAS to the negative rail.

This is simply to maintain bias currents which would otherwise rise with increased rail voltage.

That's 2 resistors and 5 or 6 transistors, depending on the suitability of your bias generator transistor. The best choice for this is usually a T0126 style transistor like BD139 but it is not critical at only a few volts, so ratings are not important, just good thermal (not electrical) contact to either output transistor case or the heatsink, which is easier but not as effective.

A little more work and it will be sounding better ,

[briansune]

YS thanks IAn the ideas is quite good to unchange the components and replace the push pull transistors, cause i found out that the 3055 &2955 is not the things that i expecting after few hearing. i thought it is because the transistor have a little bit burns out. But reply to your post i think the reason have comes to the ending, which is the 2n3055 is quite harsh(I think you know what i mean the sound base does not have any taste in it.)

Some people told me that metal case transistor is better than plastic one, in this part i'm not that sure, is that true? cause i have heard the 2n3019 and 2n4033 transistor and only quite frequency have different, but not much. What do you think?

[Ian Finch]

Hi briansune
You ask a complex question. As far as TO3 power goes, most inexpensive types are very old technology and so are not very linear devices and have low gain and frequency capabilities. Audio quality suffers generally but some designs seem to avoid the harsh sound better than others. Their best quality is their proven reliability in certain power applications - not usually audio though.

I believe that amplifiers using multiple output transistor pairs of any audio type can still sound good, especially at modest levels where linearity is a much easier task.

As we go down in chip dimensions and power, the difference may not be much at all in some audio applications as say, BC109 compared to later BC549. 550 etc. which we still use, even though better specified parts have been available for decades.

I believe that people attach quality to old, solidly cased and expensive components because they look better and served grandfather well. Personally, I know this is not a valid reason and proven, superior parts with more suitable specifications, easier mounting and good prices are widely available.

Unless we are attempting to build artworks, I think we should stay with modern, affordable and widely available parts which can still be contrived to produce the "old-time distortion" if we wish it. It can't be easily done vice-versa, though.

To see what most guys here on DIYaudio think are are suitable "throuh-hole" or leaded parts, check out Greg's Website which I referred to earlier. He did invite others to contribute and the list is good with clear application categories and easy use for anyone sourcing parts for projects here, via international sellers but you may have other sources with different parts. It is instructive to download free datasheets for yourself and compare the important qualities like Vceo, Ft, Cob, Hfe linearity as applicable to the different roles for audio parts. You can learn much better with facts that you understand than opinions too!

Finally, I should remind you about bias setting. Sound will be awful from any linear amplifier like this one, unless you get bias up to ~25 mA through the output stage, per output transistor pair. Class AB amplifiers require this for optimum low THD but many DIYs prefer overbias to get progressively more class A operation at the cost of more heat with the likely need to redesign the controller and also introducing another, lesser distortion. Make sure you get this right first - I ask because you don't refer to it in your amplifier.

You can navigate to the front page and the transistor spreadsheet from here;
http://users.tpg.com.au/gerskine/gre...ip%20sc480.htm

[briansune]

Hey Ian Have you ever heard about the PNP type of 3055

PNP3055 datasheet and application note, data sheet, circuit, pdf, cross reference, pinout | Datasheet Archive

Will it possible for replacing the MJ2955?

[Ian Finch]

Heh,heh - nice try briansune but pnp3055 is an N-channel MOSFET!
IIRC, there is another part called MTP 3055 and these are intended (it seems) to replace 2N3055 in some applications where the identity is similar to the original.

No, you are stuck with MJ2955 or the slightly better TIP3055/2955 pair which I prefer for audio as they have slightly lower distortion and easy TO247 mounting.

There is still a big range of parts that are not much better than 2n3055 and perhaps you know better what is available in your location. If you have access to RS components, Farnell/Element 14, Mouser or Digikey etc. the slightly higher cost of modern parts is easily offset by the much improved quality. Do you have access to such suppliers?

[briansune]

ha ha Ian you are really smart, i usually brough components in RS. Cause the component is free sent to my location.

But what's the point to not replacing the MOSFET transistor, cause i had checked the datasheet which is quite same as the TTl one, for example the Vceo Vcbo.

[Ian Finch]

OK. I guess you are not aware that FET and Bipolar Junction Transistors (BJTs) are fundamentally different solid-state technologies. You need to read this for yourself and then check the standard bipolar (like SC480) compared to MOSFET designs. This will show how the output stage differences are met. You can start with Google and Wikipedia and then read some schematics in threads here.

Sure, you can have similar maximum ratings for Vds,Vce but that does not make them operate the same way or make up for very different bias requirements, input capacitance, linearity curves and lower transconductance. If you want a Mosfet amplifier, you need to start a new PCB with a redesigned output stage to suit the particular FET type.

Unfortunately, PNP/MTP3055 are switching types so are unsuitable for audio. Lateral Mosfets such as Semelab's Alfet or Exicon brands or linear V-MOSFETs such as IRFP 640/9630 are among the relatively few Mosfets that are OK for audio.

When you have the basics of solid state electronics, choose a good book on audio electronics in your language and study. Bob Cordell's and Randy Slone's amplifier books discuss Mosfets and are widely available but I have no idea of translations or E-book availability, which could be better for you. Few people will ever learn enough to be competent but it is important to have a good, basic understanding of the parts we use in DIYaudio.

We are well off-topic now. Electronics and different designs should be left for another thread. I still think your best option is to use a 2SC 5200/2SA 1943 or FJP5200/1943 etc. output pair.These are cheap but offer an immediate drop in THD. Note that your driver pair must have plenty of current capability to drive 2N3055s which have awful gain linearity, falling to about 15 when they are working. Nope, 2N3019/4033 will not do here. They will soon die. MJE243/253 or KSC2690/A1220 are much better drivers for lazy old output transistors.

[briansune]

OKs that's all thanks a lot

__________________
regards