Tuesday, June 16, 2015

I Was Bored: The Amp & Preamp

I've been feeling very, very bored.  It hasn't helped, that Reilly has been a bit grumpy with me lately, and hasn't seen fit to discuss his thoughts on the steroid fueled sociopaths which normally are the topics of our mindless prattling.  Reilly's silence has left a bit of a void in my life.  So, I thought I would briefly branch off into a different subject that interests me.

Over the years, I've gradually had to accept the idea that most of the subjects that draw my attention are rather tedious, pointless, and perhaps less than ideal for maintaining my mental stability.  My OCD tends to steer me in directions that probably serve no practical or useful purpose, yet I can spend a ridiculous amount of time trying to better understand the subjects that pique my curiosity.  I have an unfortunately large number of these sorts of hobbies, and while they are all superficially quite different, I think they are all united by an underlying need to drive myself crazy.

One of these areas of obsessiveness is the subject of somewhat exotic home stereo equipment.  I think this fascination all started when I was a wee lad, and a friend of mine dragged me around to some of the more bizarre shops in my town, exposing me to the dreaded realm of the audiofools.  It's a strange world, that is heavily driven by outrageous marketing, targeted at fart-sniffing jazz aficionados who think that Citroen makes respectable automobiles.  Almost certainly, 90% of it is pure bullshit, but the remaining 10% can be quite nifty.  Like many of the things that interest me, there are frequent uses of the saying "it is more of an art, than a science", when describing why one piece of equipment was supposedly better than another.  That's an attitude that always irritates me, and probably launched me in a rather questionable direction, trying to better understand how these gadgets actually work.

After building an impractical number of speakers, fiddling with the design of crossovers, and discovering an irrational love of soldering, it seemed like it was time to add some new and unnecessary complications to my life.  I thought I would start mucking about with amplifiers.  In retrospect, this all seems a bit foolish, since I only own one CD, which is just a collection of Hawaiian crooner Don Ho's greatest hits, but sometimes you really need Tiny Bubbles to sound its absolute best.  Yes, this song is constantly playing on repeat at my house.  It is absolute perfection.

For one of my more recent projects, I was really feeling drawn to the idea of building an amplifier that used lateral MOSFETs in the output stage.  Santa Claus probably finds himself perpetually bombarded with children expressing their passion and desire for lateral MOSFET transistors.  They really make an ideal stocking stuffer.  While these output devices have their strengths and weaknesses, I have to admit that my main reason for going in this direction was that the available supply of these types of transistors is gradually becoming more and more difficult to get a hold of, so there was no time like the present to build something that utilized them.  When hospitals stop administering electro-shock therapy, I'll probably be the first in line to experience that too, before the opportunity is gone.  That's just how I operate.

Since Rod Elliott (ESP) offers PCBs for his P101, which is designed around a lateral MOSFET output stage, this simplified things quite a bit.  After fiddling with his design in SPICE, the design of his boards seemed to be capable of providing good results, and appeared to offer the ability to make additional modifications further down the road.  So, another adventure began, to build a simple class AB two channel amplifier.  (Sorry for the quality of the pictures, but photography isn't one of my skills).

Okey-dokey, let's talk about some of the technical nonsense that will probably only interest maybe 3 people (that's probably optimistic).  The power supply in this amp consists of an 800 VA toroidal transformer, with 40 volt AC secondaries.  This gets fed through dual rectifiers, followed by a capacitor bank that has 27,200 μF of capacitance per rail, after which it is putting out about 57 volts +/- DC.  Some people like to get excited about truly obscene amounts of capacitance in their power supplies, but I haven't been convinced of the benefits of this so far, and feel the amount of capacitance here is already more than adequate.  It's biased to about 30 mA, though I occasionally screw with this for no apparent reason.  The DC offset on each channel is about 4-8 mV.

As things currently stand, it should be putting out about 130 watts/channel into an 8 ohm speaker, with a fair bit more available into 4 ohm speakers, somewhat depending on the amp's ability to remain cool.  So far, overheating has never been an issue, and it has only gotten slightly warm even after a fair bit of abuse.  That amount of wattage may not sound like much after browsing the aisles at Best Buy, but I can comfortably say that the typical Marantz, Sony, Denon or Yamaha receiver that claims to put out 100 watts per channel is in fact probably only putting out about 1/3 to 1/2 of their claimed power in reality.  Plus, this amp has the advantage of being heavy enough to be used as a rather deadly projectile, if that is ever required.

Actually, if we are going to talk about overkill, the transformer in this amp is way more powerful than is really necessary.  The only reason why I did this is because I plan to eventually add more output transistors, for more current carrying capability.  At this point, there are four Exicon lateral MOSFETS per channel. but I'm contemplating doubling that in the near future.  It won't have a significant effect on total output power into 8 ohms, but it will increase the amps ability to deal with extremely low impedance speakers, not that it currently has run into any problems in this area.  These plans for possible, and largely unnecessary upgrades, are also why the heat sinks are a bit larger than they probably need to be.  Because of the size of this transformer, the amplifier also requires a soft start circuit (also from ESP), which is powered by a separate 10 VA transformer which is tucked away in the corner, in order to avoid tripping the circuit breaker in the basement.  There is also a separate circuit for protecting the speakers from any accidental 'Oopsies!', that is mounted on the rear wall of the case.

While there is a passive volume control built into the amp (which was ripped out of an old Marantz receiver which I brilliantly blew up while doing some tinkering), this can be bypassed by a couple of toggle switches (because toggle switches are cool!).  This allows me to have the option of controlling the amp with a separate preamp, which leads us to....

There really isn't much to say about the preamp.  It's an extremely simple P-88 (again, a design of Mr. Elliott's), that revolves around two Texas Instruments OPA2134 op-amps, with DIP switches to adjust the level of gain that the devices produce.  The preamp will also adjust balance, and switch between 4 different pairs of RCA inputs, though I put these in with the intention that they can be reassigned for other purposes later down the road.

The power supply is a simple 30 VA transformer, with 9,400 μF per rail, and held to a steady 15 volts DC +/- by LM317/337 voltage regulators (mounted on adorably tiny heat sinks).  Again, it's very simple, while simultaneously being significantly more than is required in almost every conceivable way.  While there's a fair bit of space left inside of the case, some of that should eventually get filled with additional pointless gadgetry, most likely a tuner.  Yes, I set this space aside for future bouts of boredom.

The volume control is a 10k logarithmic dual gang potentiometer made by PEC.  Most people will say that dual gang logarithmic pots can be troublesome, because of potential matching errors between the two channels.  I can't really disagree with this, because it did have some annoying irregularities when I first tested it.  Still, after a fair bit of tinkering, I managed to get the output of the left and right channel matched to within 3%, which is about as good as I think I could realistically hope for, at least without building a stepped attenuator.

With both the amplifier and the preamplifer, the main body of their cases are made of ipe wood, which was left over from work on my father's deck.  If you have never worked with this type of wood, it is just about one of the hardest woods you are likely to encounter, and is nearly indestructible.  One of the advantages of this is that I didn't need to seal the wood in any way, in order to protect it.  Compared to the amp, I probably gave slightly more thought to the woodworking in the preamp, and joined that together with 3/32" finger joints.  Honestly, the cases took significantly more time to construct than the actual electronics, even without any serious attempt at making them particularly attractive (though I do like their appearance).  Some of the other odds and ends, such as the knobs, were made from random bits of oak, walnut, or cherry.  If a random bit of wood was lying around from an old project, I tossed it in there.

The lids for both cases were made of 0.220" thick polycarbonate sheets, primarily because I thought it would be a shame to hide the guts away from view.  While drilling the holes in this, in order to provide some ventilation, I quickly discovered what an exciting material this really is, particularly when it explodes, sending dense pieces of high velocity shrapnel flying in unpredictable directions.  You live and learn...hopefully.

I can also say that my local community contributed something to these projects, even if they are currently unaware of having done so....

Yes, the aluminum sheeting used for the base in the amp and preamp came from a nearby roadwork sign that read "Speed Hump Ahead".  I ran across it one night while walking Reilly, and thought that the thickness of the metal was perfectly suited for the project, promptly liberated it, and cut it down into smaller pieces.  So, some of your tax dollars may have gone into something that actually works, which I think is probably a more positive outcome than you can normally expect.

Was it all worth the effort?  Hmm, that's a bit debatable.  None of this is a particularly sensible way to spend your time when you really think about it.  On the other hand, these sorts of projects tend to be surprisingly inexpensive, and the real cost tends to come in the number of hours you spend trapped in your basement/torture chamber.  Since I tend to spend a lot of time down there anyway, that wasn't a huge inconvenience.

How well do they perform?  I'd say that I am quite happy with them, not that this will stop me from tampering with them in the future.  The preamp performs flawlessly, though that's not surprising since it's job isn't terribly challenging.  As for the amp, it reminds me somewhat of my old Hafler DH-200 (which I also blew up in the past year, through an act of s̶t̶u̶p̶i̶d̶i̶t̶y̶  bold exploration, and then had to rebuild).  The new amp is more powerful than the Hafler, perhaps a bit cleaner sounding, and maybe has a few more bells and whistles, but I wouldn't say that the differences are particularly shocking.  I guess, in the end, I don't really subscribe to the idea that amplifiers are capable of performing magical feats of wonder, though I do think some speakers do benefit from certain types of amps.  They're simply devices for providing increased levels of voltage and current, hopefully without introducing noise (they aren't made with unicorn blood).  This amplifier does that rather well.

At this point, one of the next projects is probably going to be to throw together another two channel amp based on the Texas Instruments LM4780.  That should end up putting out about 120 watts/channel into a 4 ohm speaker, and about half of that into an 8 ohm speaker.  That would be quite a bit less power than my other amps, but still perfectly reasonable for most purposes.  I'm not sure why I would need to construct this, but I suppose I've interpreted my current failure to electrocute myself as a sign from above that I need to try harder.

While I'd say that these kinds of projects can probably outperform much of the equipment that can be found on store shelves, the extent to which that would matter to most people is probably miniscule.  Still, as hobbies go, it's not a bad way to distract yourself.  If there is a real benefit to doing these sorts of things, it probably lies in one of two areas.  For one, if my house is ever burglarized, none of my electronics will probably be deemed to be worth stealing.  Secondly, there is a reasonable possibility that something I constructed could live on after I have died, and eventually burn down some stranger's home, which is a thought that I find oddly amusing.