Serious Blocking Distortion

[Littlewyan]

Hi Guys

I have built the circuit in this link: http://redcircuits.com/Page52.htm

And it gets pretty bad blocking distortion when you go above half way on the volume control. On full volume its so bad the amp cuts out and then comes back. Now I have tried just connecting my guitar to the 2nd JFET in the preamp and that works fine, so I think its probably the Pre Amp driving the Power Amp too hard. Any ideas? You can hear the blocking distortion start to come in a bit with the volume at half way by the way and its not the speaker as I'm driving a Marshall 2 x 12" speaker cab.

Ryan

[J M Fahey]

Blocking distortion is normal and to be expected in that circuit.

Red Circuits are TERRIBLE , and that's being too kind.

Use that preamp and PSU (such ad it not being a total loss) driving a chip amp such as TDA2050 .

Use the single supply amp example from the datasheet.

[Kaz Kylheku]

The base of Q3 depends on the output voltage.

It is biased with a voltage divider that originates in the output, and is grounded through the speaker. The resistors are 470K/33K, so the output has to be at around 10V for Q3 to get 0.7V.

There is a feedback mechanism at play in that if, say, Q3 is cut off because the output voltage has dipped to low, then no currrent flows across R11 and R12, and so the whole VBE multiplier is raised toward the positive power rail. In so doing, it turns on Q5 and cuts off Q6.  But if Q5 turns on and Q6 cuts off, the output voltage has swung to the positive power rail also, and so that must turn on Q3.

For Q3 to be turned on at 0.6-0.7 VBE, the output has to be at around 10V, since the voltage divider if 470k:33k.

Perhaps there are episodes of C10 getting charged up, and then forcing the output voltage down so that Q3 is cut off when it shouldn't be.

Bang it up into a schematic and simulate ...

[Roly]

simulate

I put the power amp into LTSpice, and to cut a long story short, it's pretty 'orrible.  It has a number of problems such as asymmetric clipping, a large slew in operating point under overdrive, but one of the more curious "features" is that the gain of the OP stage is a sensitive function of the volume control position.  Some kind of instability with the volume control near maximum position, even with the preamp input shorted, seems almost certain because the main amp gain goes very high at this point - it's a bit of an abomination actually.

[teemuk]

one of the more curious "features" is that the gain of the OP stage is a sensitive function of the volume control position.

Well... It is an inverting amp so the gain is basically defined as Rf / Rin.



Needless to say, Rin in this particular design isn't constant but depends mainly on potentiometer's setting. With volume up all the way Rin is - in theory - zero and the gain jumps through the roof. I'm not surprised that the amp also becomes somewhat unstable at that point.

Red Circuits are TERRIBLE  , and that's being too kind.

This man speaks the truth. Crappy, inreliable, unstable and dangerous designs that seem to require a complete redesign and rebuild to offer even mediocre performance. IMO, they are not worth the time to get them "fixed".

[Roly]

I couldn't get the sim to do the blocking bit, but the large slew in DC conditions alone is a powerful caution.  After I noticed the undefined gain I couldn't really be bothered even doing a demolition job on this one, as JM says above...

All you need to do is combine this power amp with that bloody FET preamp full of trimpots and you would have a socket-to-socket disaster (perhaps all powered by a noisy SMPS).   

I had a look around that site and it's lousy with such abominations.  Publishing circuits like this is somewhere between a shame, a crime, and a danger to mental health.   


Now here is something for @Littlewyan that looks like it might actually work as intended;



Diff pair non-inverting input, gain internally defined by R4 & R10, moderate voltage rails and 8 ohm load within the capabilities of 2N3055/2955, conservative 40W output rating.  For simplicity all other transistors could be BD139/140's.  Given a sufficient heatsink this amp should actually be capable of a bit more than 40 watts.

The only thing I don't like about this design is the position of the bias setting pot VR1.  If it were to go open circuit, likely with most trim pots, the output stage would be grossly over-biased and no doubt damaged.  The right place for this adjuster is at R8.  A few fuses wouldn't hurt either.

[Littlewyan]

Thanks for everyone's feedback regarding this. I only built this amp as it was 10watts, so not too loud to use at band practice but its turned out to be a right pain. I should add that I'm also using a regulated PSU for this that only puts out 24volts, now as hes using an unregulated PSU his will probably provide 33volts under no load and maybe 28volts under load, could this be the issue? Also the blocking distortion starts with the volume halfway and sounds awful. I looked at that Chip but it puts out 32watts which is a bit too loud for what I want. Anyone know of a chip or design that only puts out 10watts at 8ohms?

[J M Fahey]

Well, that's a common misconception, so let's clar it.

In fact, the power your speaker gets comes from the PSU and nowhere else, and the amp is merely the "faucet" or "trottle" that lets that power get to your speaker in a controlled way.

So if you built a PSU for a 10W amp, and you connect a "32W" chipamp there, guess what will your speaker get?
a) 32W
b) 10W

If you guessed 10W .... you won.

So go ahead with it.

EDIT: I forgot:

I couldn't get the sim to do the blocking bit, but the large slew in DC conditions alone is a powerful caution.

That's because you applied a polite , constant, well mannered sinewave.

But apply to it any real World music signal with 10 or 15V peaks, followed by, say, 100 to 500mV RMS music (Guitar/Bass/Hip Hop/Tecno/whatever) and look what happens.
Don't know whether the Sim program is sophisticated enough to simulate this, so I'll save you the waste of time: the power amp will block to death, the speaker out will stick to one rail for a few seconds, and will slowly come to normal while horribly distorting the signal.

Those seconds, specially if you are onstage performing before some people, will be the longest in your life.

How do I know?
Well, a friend of a friend of a friend told me there was this 16 Y.O. "whiz kid" around 1969, trying to design his first SS amps, and he had *big* trouble with this      

Fact is, the transistor base makes a nice diode together with the emotter, which to worsen things is directly connected to ground, so on signals over 700mV peak it charges the coupling capacitor real quick.
This reverse biases the BE diode, so only discharge path is through the base biasing resistor.

Now do the Math, with 47uF (C9) and R14 (33K)
I'll save you some time, the RC time constant above is 1.5 seconds.

How would you like having random 1.5 seconds muting going on and off in the middle of a hot solo? 

[Roly]

JM> "polite", "well mannered"

 

Not me.  I gave it a nasty flogging,    but it's a sim, and the best sim is the real world where component models aren't constrained or idealised, or "off Broadway" operation skimped or ignored, so once I saw about 10-15 volts of Q-point shift under drive I just couldn't be bothered following a lost cause any further.  I may not get it to actually blocking-oscillate in the sim but I have no doubts that it does so in real life.  As a sidebar, LTSpice has the ability to both import sound files as circuit drive, and export resulting sound files from sim circuit output.  Neat!   

You are quite right that these is a lot more "dog's breath" on that site; noticed a "150W" amp running a single 2N3055/MJ2955 pair on +/-42V, never mind the device datasheet (cough splutter).  The worst part about this sort of idiocy is that exactly the sort of people who will get caught, newbies, will often assume that it's all their fault they have a cloud of smoke and a pile of dead devices, can't seem to get it to work, yet it's a published and "recommended" design.  It's a pretty awful realisation that there are people around who will swear blind on their sainted grandmother that they built it, it delivers 150W_RMS measured, and that it has been delivering awesome levels and tone at gigs for years - when it's entirely a wind-up, not a word of truth in any of it - all bulldust.   

Red makes play of featuring in the Reader Circuits section of Electronics World (was Wireless World, UK).

Well, as it happens I have some personal history here having been a lifetime reader of WW until it became EW and the standards dropped through the floor.  Between then and when I finally gave it up as a lost cause I had quite a number of e-mail exchanges with Svetlana (the editor, not the valve) about technical issues, mainly arising from these reader contributed circuits which variously wouldn't work, would catch fire, or in one case an electric fence that would kill somebody.  The implication on Red is that their circuits have been subject to some sort of peer review by being accepted for publication, but in reality nothing could be further from the truth.   

{I just wish I could do a hot solo for 1.5 seconds.  }

[Littlewyan]

J M

Sorry to maybe sound a bit dumb, but can you explain what you mean with regards to the PSU? My PSU provides 24volts up to 2A, does the voltage determine the power output? As this chip can take up to 25volts, so would I have to lower my voltage a bit?

Ryan

[J M Fahey]

There's no dumb question 

To answer with more precision, please post what chipamp are you talking about.

I had *assumed* it was a TDA2050, since it's the only "32W" popular chipamp I know, but since it stands way over 25V you must be talking about some other one.

[Littlewyan]

Well the TDA2050 datasheet states that the supply voltage is 25V? So does that not mean it would push out 32watts at 25volts?

[blackcorvo]

Well the TDA2050 datasheet states that the supply voltage is 25V? So does that not mean it would push out 32watts at 25volts?

It actually says +/- 25v, which means a symmetrical supply of +25 and -25v (50v).

Since you're looking for not-so-loud amplification, you could use a TDA2030 instead of a TDA2050. It's used in most 10W guitar combos out there, and it can handle 24v without any problems (max voltage for it is +/- 18v or 36v).

[stormbringer]

Just a Little curious about one thing, just to make sure. when you say your supply is 24v, does that mean single or 2x12 with a center tap referenced to ground?

[Littlewyan]

The transformer is a 2x30V, center tapped to ground and goes through a variable voltage regulator that is currently set to 24 volts. I only decided to use a regulated supply as I've never built one before!