newbie questions about transistors and power suplies

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I hate being the newb, but no better way to learn than to ask :)

question 1: Why is a torroid transformer used in all the power supplies I see on this board and liked to from theis board?
I built a qkits 100watt amp a while back and used the iron laminate transformer out of a PA amp in the power supply. this worked fine until It met a 2 ohm load and smoked.. still no idea how to fix it :( does the torroid produce less noise or something? I thought the whole idea of laminated cores was to reduce eddy currents and their waste heat.. ??

question 2: Is it possible to make a MASSIVE power supply that merely rectifies AC right from the socket for +/- 170 peak?
And if you do this, can you make an amp that uses high voltage transistors, but only swing them maybe +/- 40 volts?

question 3 closely related to question 2 : If you operate an output transistor(s) at low power levels, like 5% if your rail voltages, do they have to dissipate 95% of the rail voltage?

sorry if these are really dumb questions.. Im a computer programmer, so only know enough about electronics to be dangerous.
Addressing question 2 only:

Rectifying the power from a wall outlet would really be nice for power amps but it's to dangerous for a commercial product.

A DIYer could do it for their own system but they should be aware of the lethal potentials. Don't fry your kids or guests.

The power amp could have input transformers for isolatiing the pre-amp and signal sources but the amp and speaker connections would still be HOT. I don't know if really good input transformers are available. A digital optical link is an alternative.

As to the availability of suitable linear, high voltage solid state devices for the amp, I couldn't tell you. Start seaching the data books. Maybe Nelson Pass has some info.

But, what's the real point except for situations that require kilowatts of power? Personally, I don't think it's worth the trouble just to save 20 pounds of weight and $100.
i found kilowatts thread about the 240's mains rectified 1800 watt amp he was pondering. One of the threads pointed out that with striagt rectified AC the floating ground would be at 170 volts over mains ground... the transformer in a normal power supply lets you tie the float ground to real ground and end up with true +/- rails with respect to ground. I understand now the issue with the ground and the input signal ground being off and the safety issues with that.
So how about a switching power supply using straight rectified AC ??

Also , really want to understand question 3, can someone help??
Actually the amp I was pondering would plug into a 120/240V 50A outlet, so there is 2 120VAC legs and a neutral, just like a center tapped transformer. You get a +/-170V supply. Quite convenient. There is no ground issues. It can be unsafe as a commercial product because there is no isolation from the line, so the inputs must all be isolated and the speaker outputs must not be exposed.

The reason I changed the project's design is because there are no efficient 900W 8ohm speakers, and you don't get good sound if you put speakers in series at those levels. And (the main reason) if you have a 7200W system, you want it to be redundant anyway, like for PA or something (what else would you use a system like that for?). So I am doing lots and lots of lower voltage stages. You don't want to have a single amp failure stop the show.

An amp with over 160V rails would only be practical for maybe ES speakers or something, and the channels are very hard to design because of slew rate and stability.

Question 1: Toroids minimize radiated EM interference. They make your amp quieter. And they look cool.:D

Question 3: In a class A amp, there is always a large current flowing through the output stage, the devices are always on. In a class B or AB amp, if there is no signal applied, the output devices have the highest possible resistance across them, and little current flows. As the signal increases, more power is allowed to flow through the sum of the output stage and load. But at lower signal levels, and therefore lower output, a higher percent of the voltage is dropped across the output devices, so during a peak near the rails, the amp does not dissapate as much heat. If the signal is at 5% of the rail voltage, 95% of the rail voltage is dropped across the output stage, but is the power output is low, this will not be that much, but the amp will not be very efficient. If you wanted a 400W into 4ohms amp with 170V rails, it could double as a central heating system. :confused: I hope this helps.
okay, that was exactly the answer i needed.
maybe you can help me with this next question:
I built (about 4 years ago) an amplifier it from Maplin, model LP33L that I canot find anything about on the internet. I blew it on a 2 ohm load accidentally. (didnt realize it was a 2 ohm, was told it was 8 ohm)
After finding this forum I am ready to dust it off and hopefully with help learn how to figure out what is wrong with it. I supose I will need to scan the documetns I have for it and upload them, since I cant seem to find anything about Mapln online... (did they disapear? I bought this off of ??)

anyways, it is a "bipolar transistor based design with a complementary push pull (class b) output stage" 100 rms @ 8 ohm 175 rms @4 ohm with a +/- 44 volt power supply.

What would be the standard approach to figure out what is wrong with this amp? I figure the output transistors are blown... is there any simnple way to test? All i have is a meter that can do capacitance and diode check and standard volt/amp/ohm.
If I get the parts list and schematic and stuf posted, could someone help me??
First, measure the resistance across the C/E of the output devices. If you read a short, then they are fried, this is a common type of transistor failure. If they are open you will not know unless you make a simple test circuit to see if the transistor operates, or get a transistor tester. The drivers may well also be toast. But I personally am not the best guy to ask about repair.

Oh, and if the outputs are bad, I would replace those 2N3055/MJ2955 pieces of junk with MJ15003/MJ15004 devices. They have higher power dissapation, higher voltage and current capability (+/-44V rails are pushing it with 2N3055/MJ2955 devices, they are rated 60V and must withstand 88V in your amp), better thermal junction to case contact, and are better matched. They also sound better.
They should work fine as a drop-in replacement. All TO-3 power transistors have the same pinout. Just be sure you get the correct NPN/PNP devices on the proper sides of the circuit.

Do you have schematics or anything? It would be nice to have a look before you begin.
No Mains transformer and more stuff

question 1: Why is a torroid transformer used in all the power supplies I see on this board:

Because of several things I guess

1: There is almost no external magnetic field around a torroid, this is a very nice thing since magnetic low frequency fields are hard so shield in. This results in that the wirering around the transformer is less critical.
2: The are easy to mount, because it's most likly on hole that needs drilling.
3: Historical reasons, and religion in HIFI demands a torroid beeing used. (BTW: any transformer with the right V*A ratio will be fine to start with)

question 2: Is it possible to make a MASSIVE power supply that merely rectifies AC right from the socket for +/- 170 peak?:
YES !, but take great care !!!!!!!!!!!!!

I think i saw it done in a SUNFIRE active subwoofer, but i don't remember.

question 3 closely related to question 2 : If you operate an output transistor(s) at low power levels, like 5% if your rail voltages, do they have to dissipate 95% of the rail voltage:

That is a guestion of what kind of outputstage you use (CLASS G and H are often used to get a smaller Heatloss), But genneraly the answer is yes, I you come up with an amp where this i not the case please let med know :)

Hope it helps

i have the parts list and the schematic within a 4 page build guide. I dont have a working scanner on theis PC, and it is 2 am, so it will have to wait til tomorow. I will get those scans, pics of the assembled mess, and the results of the transistor test you suggested here tomorow afternoon.
here is the scan of the pamplet

Transistors have direct short from any terminal to any other terminal. This is in circuit (okay or need to measure out of circuit?)

If all I do is buy the MJ15003/MJ15004 pairs to replace the 2N3055/MJ2955 pairs it should work as good or better than original? (assuming it is transistors that are only thing bad?)
interesting... when searching for datasheets on the MJ15003 i found a web page about conterfeit parts... well the pic has a 2n3055 in it, and the 2n3055 that came in my maplin kit bought from has an almost white top, with black lettering, oriented the same way, but with an ST logo, and the print is almost gone !! I think it is a fake.. I am gong to open it up since it is dead anyways.
To start:

If you have to ask the question, you don't have enough expertise to try it.


The risk of electrocution and fire is too great.

Transformers are your friends.

I do not know that a toroid tarnsformer is automatically better. You can point to some specs and support the argument. Like many things in engineering/design it is a series of choices and trade-offs. Toroids are popular and do have thier advantages.

In a power amp, I use them. However, there have been plenty of well regarded commercial power amps (from well regarded engineeers) that used EI transformers.

In pre-amps and lower level ciruits, EI laminates have thier advantages.

The transistor answer depends on which amplifier class you are running the output stage. Class A, your 5% - 95% concept is nominally correct. In Class AB (some texts call it this Call B) you do not have this "extra wasted power."

The "extra wasted power" gets disapated in the form of heat. Class A amps need large heatsinks. Class AB run cooler and do not need as large heatsinks.

My above explaination is simplified. It is more complex than I have stated above. It will help you make the big decision about what kind of output stage you wish to impliment.

I would suggest 2 books:

High-Power Audio Amplifier Construction Manual
by G. Randy Slone, Randy Slone

Audio Power Amplifier Design Handbook
by Douglas Self


This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.