Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 02:00:42 PM

Login with username, password and session length

Recent Posts

 

Serious Blocking Distortion

Started by Littlewyan, July 05, 2013, 10:54:13 AM

Previous topic - Next topic

Roly

Quote from: LittlewyanI'll stay clear of RedCircuit in future

Wise move.  I try to give credit where it's due, but after reviewing a number of the circuits in all sections on that site it really is a large collection of carefully selected rubbish.  In fact it's almost amazing how few workable circuits there are there.

Quote from: LittlewyanWhat were the two diodes for by the way? Distortion?

No, it's intended as a clean amplifier.

The primary function of these two diodes is to prevent the amp output going above the supply or below ground where it might be damaged.

The reason that this is a concern is that a loudspeaker is not a pure resistance but an inductance which can store energy, and in some situations it is possible for this stored energy to be returned to the circuit in such a way as to drive the output above or below the power supply rails, and these diodes are there to prevent that and protect the chip from damage.  In many situations they would not actually be needed, but datasheet circuits tend to try and cover every possibility, no matter how remote.
If you say theory and practice don't agree you haven't applied enough theory.

Littlewyan

Hi Guys, I've drawn up schematics for both the PSU and the Amp itself and attached them to this post. What do you think?

phatt

@ Littlewyan.
Your  understanding of PSU needs work.
Ask yourself,,What does the bottom Diode Do?????????

Can I suggest;
http://sound.westhost.com/project27.htm (down the bottom is PSU)

http://sound.westhost.com/index2.html
Go to the site map for tons more stuff. You will be there a long time so take lunch ;)

Or Dload Teemu's PDF Book on SSAmps right here on SS Guitar. winky.

http://www.ssguitar.com/index.php?topic=711.0

Phil. 

Littlewyan

Does this not increase the load capability of the PSU? As one diode can take 1A, so surely one on each winding means i can draw 2A?

The first link you sent me btw was a PSU design i planned to use in future, looks much better ;)

Roly

PSU;

- The cathodes of the rectifier diodes should be connected.

- The 1uF's would be better as 0.1uF's, and right on the regulator pins.

- The voltage adjustment is a bit fraught; if the wiper goes open on the track the output voltage goes to the unregulated max.  It is better to use a fixed resistor that is a bit larger than required to give an output voltage a bit higher than required, then trim it down with a trimpot in parallel.

- the 100k across the output is not required.


With the amplifier; there is no need to have the input resistance as low as 220k.  The first FET looks effectively like an open circuit, so the Gate return resistor may as well be at least 1 megohm, particularly if you will be using it with a passive guitar, and could even go as high as 10 megs for tone reasons.

- The 100nF input cap, 100pF RF stopper, and 100nF power supply bypass, and 0.47uF in the Zobel are shown as polarised, they aren't.

- There should be a bypass cap, say 100uF, on the supply rail after the 1k, on the Drain of the second FET.

- A de-popping resistor would be a good idea across the output to charge up the 1000uF if the speaker isn't connected, say 1k.


P.S. the difference between one diode and two in this case is between a half-wave and full wave rectifier.
If you say theory and practice don't agree you haven't applied enough theory.

Littlewyan

O god i'm sorry i forgot to connect them, apologies it was late last night i drew these up. Ok i'll make amendments tonight. The trim pot modification on the PSU may be a bit difficult as its already built and bit tight on space in that area but i'll see what i can do.

Littlewyan

I've amended the schematics and attached them to this post. The PSU I may leave as is as it is already built and working, however I will add the 4.7K Resistor in parallel with the Trim Pot to give a maximum of 24volts. I've also left that 100K resistor on the PSU in case I want to test the PSU without the amp connected.

With regards to the AMP the capacitors you specified are on the schematic as non polarized, they just use the same symbol as the polarized capacitors except for the missing positive sign. I will add an 100uF capacitor by Q2's Drain if need be but it does work ok with that 470uF on Q1's Drain at the moment.

J M Fahey

It looks fine and should work well.
Have you already tried it?

Only minor problem is that the tone control will not work as intended, since it's trying to shunt highs to ground from a low impdance point, but that's a minor matter.

After you check your amp is fine, I'll suggest a more useful tone control-.

Another minor problem is that those 1N400x PSU diodes are barely usable (1A), I'd replace them with 1N5402 or higher (3A).

But it's already looking fine  :dbtu:

Littlewyan

It does work quite well at the moment though? A little too well to be honest it can make the sound so bassy that you can barely hear the high notes!

This amp won't use more than 2amps will it?

Roly

Quote from: LittlewyanThe trim pot modification on the PSU may be a bit difficult as its already built and bit tight on space in that area but i'll see what i can do.

Meh, your build; can you live with the possibility of raw power finding its way to the chip amp?  I think it's below the maximum, and they tend to be pretty rugged and ignore crud on the supply anyway...  :shrug:

Quote from: Littlewyanthe capacitors you specified are on the schematic as non polarized, they just use the same symbol as the polarized capacitors except for the missing positive sign.

They are an abomination that is neither the US nor Euro stantard symbol.



There are lots of young programmers who want to reshape the world in their own image, and while "standards are as exciting as a Russian truck" they also allow communication without confusion.

Quote from: Littlewyanit does work ok with that 470uF on Q1's Drain at the moment.

100uF or 470uF is no matter, but nothinguF was shown on your Mk1 cct.

Quote from: LittlewyanThis amp won't use more than 2amps will it?

DC? No.  But a) 1N400x are 1 amp diodes, and b) the peak current that flows on each cycle of the mains to recharge the filter cap is very considerably higher than the DC or average current.  Diodes are rated with this is mind, however that doesn't change the fact that the 1N400x-series are only one amp (average) diodes.

Quote from: LittlewyanA little too well to be honest it can make the sound so bassy that you can barely hear the high notes!

Which isn't surprising considering that at the extreme it simply connects 1uF across the signal path - a bit savage.  I'd lose that tone arrangement entirely and look at putting a more conventional tonestack in  just ahead of your "Gain" control.  Failing that you could reduce the 1uF to something more like 0.1uF.
If you say theory and practice don't agree you haven't applied enough theory.

Littlewyan

Alright I didn't mean to upset anyone by picking the wrong symbol, I just used that symbol as it was there in front of me and I understood what it meant. If I was publishing this schematic online (which I most definitely will not) then I may of put a bit more thought into it. I'm not a young whippersnapper trying to reshape the world, I'm just trying to learn about electronics.

It wasn't on the schematic you're right, I just forgot to add it in, that capacitor is currently in the circuit. I'm not saying you're wrong in anyway, I was just pointing out the circuit currently works with the 470uF in it (well barring the power amp obv), sorry I forgot to point out my mistake.

I will add that trim pot modification onto the PSU, just need to work out how due to space etc. It may be below the maximum for the chip but it won't do my Pre Amp much good as those Transistors can only take up to 25volts. Not sure how the guy who originally designed the amp had 33volts going on it as those Transistors should of blown surely.

Roly

Quote from: LittlewyanNot sure how the guy who originally designed the amp had 33volts going on it as those Transistors should of blown surely.

And here we are back again at what JM and I were saying above about some of the circuit sites that are around the net being a health hazard.  They walk like a duck, quack like a duck, but are actually turkeys.


I ain't upset - it's a matter of what sort of impression you want to give/leave (and by uploading your circuits here you are actually publishing on the net).  Give Google the right keyword search and it will pull up your circuit, or plonk somebody down in the middle of this thread.

Drawings are another form of descriptive language, and be they architectural, engineering fabrication, hydraulic, or electronic, they all have conventions, and if you step outside those conventions then communications starts to get muddied.  There are some drafting programmes around that have been put together by keen and effective programmers, but some of these haven't bothered to find out what these drafting conventions are, and why, and simply invented their own.  While these work to a greater of lesser extent you frequently find that when you want to do something entirely reasonable in conventional electronic drafting you either can't, or can only do it wrong.
If you say theory and practice don't agree you haven't applied enough theory.

Littlewyan

You made a very good point Roly, so I've edited my schematics. I shall remember to use these symbols in future.

I have now ordered a couple of TDA2050s, so once they arrive I shall dig out the rest of the parts and work out how I'm going to lay out my PCB. I will also change that tone control capacitor to try and improve things, just for the moment anyway, might change it altogether as suggested by yourself.

Roly

Yo!  Now the instant I look at that it starts to tell me its story (but you still have that 100k on the output of the power supply that you don't actually need).

There is another convention which we have discussed here in the past, and it concerns how you draw wire that cross and connect, and that cross and don't connect.

There are (like caps) a couple of conventions.  One is that crossing wire that join have a dot and those that don't join don't have a dot.  The other is that wires that join simply cross, and wires that don't join have a little hop-over loop.

Straight away we can see a couple of potential confusions, that join in one convention is don't join in the other, and that a missing dot will change a join to a not join.

For 50-some years now I have followed a mixed convention; wires that join cross with a dot, and those that don't hop over.  This is neither convention, but it is also another step away from a possible misunderstanding, and in the end that has to be the dominant thing, a drawing that represents the situation in the best possible way that avoids misunderstanding.

While I may draw my circuits like that whenever I can (and some software can make it very hard to do what you want) the world is still full of people who follow different conventions so you will never entirely escape having to deal with them.

Guitar amp tone controls generally are an area where you can have almost endless fun trying stuff out.
If you say theory and practice don't agree you haven't applied enough theory.

J M Fahey

That's the clearest meaning convention, and it follows "real world" stuff inside the chassis: where wires join you *do* have a "dot": the solder blob, and where they don't. one "jumps" the other, they are definitely not in the same plane.