In US/Can parts stores, 1156 is the old-old standard "brake light", 12.7V 26.9W (2.1A)
1157 has one filament like a '56 and one "running light" at about a third the current, 8.3W, needs a double-contact socket you only find in car-parts stores.
1141 is 18.4W 12.8V 1.44A.
In their loud (for the time) 2.7" Sentry (IV??) speaker, E-V used a 28V aircraft bulb. You could see it flashing through the grille.
Get them while you can. And avoid LED "equivalents".
From before I could drive until last year I had a stash of 1156/1157 always at hand. Last year I realized that not even the 1993 Chevy pickup (RIP) and not the 1996 or 2002 Hondas used the 115_ series lamps; only the 1991 Miata has big bayonet lamps. Later cars have the baseless "wedge" lamps which tend to age badly (rusted to the socket so bad you have to get a whole socket assembly for the specific model).
Car headlights ran 35W-55W. 35W is more common on LO, but tonight two guys followed me with 55++W lights.
I checked into that one, current is too high...(2A instead of the 1A that I guessed, and Fahey confirmed).
I looked at that datasheet, too. IIRC, long ago and far away, I soldered wires directly to the three wire leads emerging from an 1157 bulb. I seem to remember there was a flat rectangular plastic base, with the wire leads wrapped once around it, each in their own individual groove in the plastic.
But once again the current limit isn't quite right, though the 8.3 W filament would work (maybe too conservative, but that's better than blown tweeters.)
I missed that one. 1.44A might be a tad high - what if I parallel a suitable power resistor across the tweeter terminals, so the voice coil only sees 1 amp when 1.44 amp total current flows? (i.e. pick a resistor that flows 0.44 A when there's one amp through the tweeter.)
Exactly! The idea (to try and use a dome tweeter in a little 30W - 40W personal monitor/tiny PA) only occurred to me a few days ago. When I started looking for incandescent auto bulbs, I found they are already mostly extinct in the auto parts stores, Amazon, et cetera.
Mebbe I can find some on Ebay. Also need to go through my junk-box. Fingers crossed!
If those don't pan out, I might try and find a local Pick-A-Part type auto junkyard. Surely there will be some good bulbs left in all those hundreds of cars.
There's still another use for tiny tungsten-filament bulbs that might be relevant to some diyAudio members: they make good amplitude stabilizers for Wien Bridge low-distortion audio oscillators.
IIRC we need a low-voltage, 40 mA (0.04A!) bulb for that. Those are even less likely to be available now, compared to filament lamps for automobiles.
The tiny NTC thermistors that were the alternative method of stabilizing Wien Bridge amplitude died out and went extinct many years ago, long before the grain-of-wheat bulbs.
That's the trouble when one technology dies. Some of us have found alternative uses for it, and the new and shiny replacement technology usually doesn't work for those alternative purposes.
In my impoverished childhood, my dad's cast-off double-edged razor blades were an incredibly useful item. Break one into two halves lengthwise, and I had two free Exacto-knife equivalents. Or a little leaf-spring for the various mechanical contraptions I was constantly building.
Magnetize a cast-off razor blade by stroking it with a bar magnet, then float it (gently!) on water, and you had a magnetic compass. (I guess there wasn't enough chromium in the alloy to prevent them from turning into weak permanent magnets.)
Et cetera. There were many more uses. All of them disappeared when double-edged razor blades went away, and were replaced by those silly slivers of metal stuck in a plastic razor-head, or by an expensive buzzing electric thingy.
For that matter, the replacement technology doesn't work as well for its intended purpose, either. Most people seem happy shaving with them, but my heavy-duty stubble scoffs at the silly things. Electric shavers are even worse; I once shaved one side of my face with one, leaving the other side unshaved. I couldn't tell the difference, and neither could my wife.
It turns out there is a sub-culture of men still using double-edged razor blades, because they get a much better shave with them. I'm one of them.
-Gnobuddy
Read again please. I said shunting the LED - not in-series with. And then there's six, paragraphed bullets spawned by your miss-read. That's OK.
The LED begins to sip well short of 1.8V and the LDR sees it. If my three devices and yours are at all similar, and I have my LDR shunted with a 10K Rf by your graph the LED current is ~low 10s of uA when the 10K begins to "cross over" b/c the LDR has fallen below 100K. Along with the LED shunt this happens around Vo = 2.9V (i.e. linear to that point). The LDR // 10K combo then follows a downward slope of some kind (i.e. there is an in-between area) ending in a few hundred ohms with Vo at that point = say 4V - just short of the op amp's rail limit.
I don't use Hollis' 2nd op amp half you call a "rectifier". There is no R7. Hollis needed that voltage stretcher b/c the TL072 can't rail well. Had nothing to do with rectification. In any case, we both agreed that his circuit is no more than a hard limiter unless he was using some Vactrol we don't know about.
I'm sure you did but did think of picking the range of response you need from your graph and adding a fixed shunt. Stay away from that units-of-uA range
PS. Not all EEs are alike. It's not special b/c you work with log-log graphs. That was a wee bit narcissistic of you.
So you posted a schematic that looks nothing like the circuit you're actually building?
No wonder I misunderstood you, my friend!
One of the reasons why I measured my optocoupler is to find out what range of LED currents I would actually need to design for. For me, for my application, that turns out to be the range from 10uA to 1 mA.
If you post the actual circuit you're building, someone reading this thread can try to help. Without it, we can't.
I disagree. The second stage is a crude precision full-wave bridge rectifier, designed to light up the LED as rapidly as possible when a signal starts, whether the first half-cycle happens to be positive-going, or negative-going.
The low E from a guitar is at about 83 Hz. One cycle lasts over 12 milliseconds. A half-wave rectifier circuit (which only senses one polarity) might have to wait that long - doing nothing - before it can even begin to react to the start of a new guitar note. There will be additional further delays after that, because of the LDR's own response time, as well as the electrical attack time constant.
This sort of slow attack will cause a loud and obnoxious "pop" at the start of a guitar note, because a loud signal gets through for 10 - 20 mS before the compressor can even begin to squash the signal.
Every compressor I've tried with guitar suffers from this exact problem. It's worst if you hold a long note (letting the compressor climb back to full gain), and then pick a new note (which will go "Pop!" before the compressor can respond and tame it).
Robert Fripp and some other famous guitarists from decades past used compression extensively. I don't know how he tamed the popping.
The schematic you posted is going to be a hard limiter because there is too much voltage gain ahead of the rectifier.
Since I haven't seen the schematic you're building, I have no idea how that will behave.
Holy cow.
I think you've colossally misunderstood me. There is nothing special about working with log-log graphs. If you've ever looked at a frequency response plotted in the usual way (decibels on the y-axis, log-frequency on the x-axis), you've looked at a log-log graph. If you've ever read "...filter slope of 6 dB/octave...", you're talking about a log-log graph.
Surely everyone who's interested in audio electronics is used to working with those?
-Gnobuddy
Yes, I never use dome tweeters because of low sensitivity and power handling, but such a bulb should protect them anyway.
Speaker design is always*a compromise, domes are for Hi Fi so extended response, flatness and dispersion are paramount, while here for Bass/Keyboard reaching flat to 10kHz is more than enough, higher sensitivity and narrower directivity are a bonus because they put "more SPL" into audience ears and the final "parameter": I *can build them in house, quickly and for low cost.
Mind you, I can make dome tweeters, but construction is more precise and delicate, must buy ready made dome+coil assemblies and really provide no advantage here.
In fact, in the heroic old days, I made full horn drivers in 2 sizes: 2" ones for PA midrange , with phase plugs and chambers turned from solid aluminum + commercial aftermarket domes, and 1" (think the ubiquitous Fostex horns) with Bakelite pressed plugs and chambers.
Even some prototype Goodmans cast horn clones
Joys of Tariff protection days, today disappeared to help build the Second Chinese Empire
I was talking the humble 12V 15W (there might be a 12V 12W version) "position light" (we call it that way) ,maybe US/UK "tail light" which neither signals something special (such as braking) nor illuminates road ahead for driver, simply tells the World "I am here".
Typically bayonet mount.
FWIW this is the actual crossover plate in a "fridge" type Ampeg cabinet, showing said lamp in all its glory:
https://www.lamps-on-line.com/car-bulb-211-12v-15w-ba15s.html
Speaker design is always*a compromise, domes are for Hi Fi so extended response, flatness and dispersion are paramount, while here for Bass/Keyboard reaching flat to 10kHz is more than enough, higher sensitivity and narrower directivity are a bonus because they put "more SPL" into audience ears and the final "parameter": I *can build them in house, quickly and for low cost.
Mind you, I can make dome tweeters, but construction is more precise and delicate, must buy ready made dome+coil assemblies and really provide no advantage here.
In fact, in the heroic old days, I made full horn drivers in 2 sizes: 2" ones for PA midrange , with phase plugs and chambers turned from solid aluminum + commercial aftermarket domes, and 1" (think the ubiquitous Fostex horns) with Bakelite pressed
plugs and chambers.Even some prototype Goodmans cast horn clones
Joys of Tariff protection days, today disappeared to help build the Second Chinese Empire
I didn´t mention or refer to brake lights, because those by definition mussy be twice as bright as regular ones, to make braking noticeable to the diver behind, so he can brake too to avoid a collision.
I was talking the humble 12V 15W (there might be a 12V 12W version) "position light" (we call it that way) ,maybe US/UK "tail light" which neither signals something special (such as braking) nor illuminates road ahead for driver, simply tells the World "I am here".
Typically bayonet mount.
FWIW this is the actual crossover plate in a "fridge" type Ampeg cabinet, showing said lamp in all its glory:
https://www.lamps-on-line.com/car-bulb-211-12v-15w-ba15s.html
Last edited:
Agreed, I would be quite happy with a nice flat response to 10 kHz. At this point in my life I probably can't hear much above 12 kHz anyway.
I remember the 1980s, when the high-pitched whistle from CRT computer monitors used to drive me nuts, while nobody around me could hear anything at all. But those days are long gone!
I'm curious to try a dome tweeter simply because they're easily available, not too expensive, and it's not too hard to find one with a decently flat frequency response. Because I don't want a lot of SPL, I think a dome tweeter might do the trick for me.
Until recently, I never would have considered this, because dome tweeters are famously delicate. The speaker engineers I worked with a couple of decades ago were firm in their opinion that dome tweeters had no place in any speaker system used with musical instruments. That opinion was confirmed by an acquaintance who blew out the tweeters on his nice Alesis monitors by playing his electronic keyboard through them.
But those engineers didn't know about the light-bulb trick. Nor did I, until a few days ago.
For a tweeter, I'm looking at the Visaton SC-10N as a possibility ( https://www.visaton.de/en/products/drivers/dome-tweeters/sc-10-n-8-ohm ).
Visaton provides a comprehensive data sheet, as usual for the brand. Frequency response is quite nice and flat from below 2kHz to at least 15 kHz.
Digikey Canada stocks these, which is a big plus for me. Sensitivity is 90 dB@1W@1m. Power rating is 100 watts in a system crossed over at 4 kHz, or 50 watts in a system crossed over at 2 kHz.
They're not exactly cheap by my standards ($32.44 CAD at Digikey Canada), but how much does a commercial small PA speaker cost, particularly one with good tweeters?
Rod Elliott's website contains a graph showing that about 12% of total power is in the band above 4 kHz, suggesting this tweeter can actually handle about 12 watts. (12% of 100 watts = 12 watts).
The same graph suggests 18% of the total power is in the band above 2 kHz. The tweeter rating of 50W with a 2 kHz crossover suggests it can handle about 9 watts (18% of 50 W).
Both numbers are close enough to the "roughly 10 watts maximum power for a tweeter" number we've heard on this thread. Which also goes nicely with a 1-amp maximum current limit (bulb).
Last night I did some more searching, and found some 12V, 12W bulbs sold for RVs, exactly for the purpose you describe. Side marker lights, etc. I think these may be the ones you recommended, or very similar: https://www.lightbulbs.com/product/westinghouse-03726
12V/12W is the "holy grail" 1-amp bulb I've been looking for. They should limit an 8-ohm tweeter to nominally 8 watts (with 20 volts RMS applied to the series bulb+tweeter).
I also found some 12V, 10W halogen bulbs with wire leads on Amazon. Nominally 0.83 amperes, or about 6 watts into an 8 ohm speaker (when 18.7 volts RMS is applied - close to the maximum square-wave output signal from the amplifier I plan to use).
With any luck I'll be able to solder to those bare wire leads. I ordered a few to try.
-Gnobuddy
Those bulbs look about perfect.
Sadly, reading comments on that page, they are becoming hard to find.
But having been SO´popular, used by the millions for decades (if not a Century by now) , somebody must still stock them.
If anything, for rusty old farmer´s pickup trucks
Hey! I would order a 10 bulb pack from China for peanuts (I hope) simply to have them around.
Both for tweeter protection replacement or even .... duh! .... car use
Remember to order a few matching bayonet sockets, unless you want to solder them in-out at every replacement.
Sadly, reading comments on that page, they are becoming hard to find.
But having been SO´popular, used by the millions for decades (if not a Century by now) , somebody must still stock them.
If anything, for rusty old farmer´s pickup trucks
Hey! I would order a 10 bulb pack from China for peanuts (I hope) simply to have them around.
Both for tweeter protection replacement or even .... duh! .... car use
Remember to order a few matching bayonet sockets, unless you want to solder them in-out at every replacement.
The trouble is that someone just sticks a few LEDs into a compatible lamp base, and sells you that instead. Fine for the rusty pickup, but no use at all as speaker protection...If anything, for rusty old farmer´s pickup trucks
These 10W, 12V bulbs are still affordable (but have wire leads): https://www.amazon.ca/Vstar®-12Volt-Hours-Landscape-Halogen/dp/B01MXYQYNL/ref=sr_1_5
Amazon Canada wants nearly $8 CAD per bulb for the Westinghouse ones. At that price, I'll look around a bit more before I decide to buy them.
-Gnobuddy
Here you go again slamming me b/c you can't read or comprehend. I'm irked now. Go back to Post #286 and really read it "friend".So you posted a schematic that looks nothing like the circuit you're actually building?
No wonder I misunderstood you, my friend!
Great minds think alike. That seems like a good region and is where I'm at "on the bench" just playing right now. OK? No JRA schemo YET.
I'm not there yet. I posted a circuit as an EXAMPLE from a guy named "Hollis". Go back to Post #286 and really read it.
The whole circuit is based on that Germanium diode full-wave bridge rectification so what's your point? He had a 9V budget, a combined Vf threshold of like what, 2.5V to overcome, an op amp that could only manage +/- 3Vpp to do that with and that didn't cut it. If the bridge was Si - no go at all. Nothing magic about Ge in this case other than lower Vf. That 2nd amp does little more than be a differential helper and the Ge could be Si. And, after your own experience with a "vactrol" do you think rapid "slam the LED" light-up is necessary, in this case? Maybe so. As I WROTE EARLIER I haven't seen that delay you see with the device you purchased. But I'm not interested in the 1uA region - and all is still on the bench doing static empirical testing.
First off, a FULL-WAVE BRIDGE is feeding the LED, Not half-wave. Keep things factual. My first purchase was supposed to be a Perkin-Elmer clone of their old VTL5C3. Albeit with certain disclaimers IAW P-E's data sheet, its 63% turn-on time was 2.5mS and with a decay of 35mS. I've done NO dynamic testing. So lay off. Grit your teeth and do not reply. Just shut up for once.
Running light or parking light. (The exact function has always been unclear- certainly we don't leave them on to park all night, but that's what we call them; running light makes it clear they are for being seen from the rear or when headlights are out.)
Where do your car electrics come from? English customs? Lucas, Prince Of Darkness? The US/Can 11__ series was stable from late 1930s (except 6V) to early 1990s. I have never seen a BA15. (I forget what the 1972 BSA cycle had; they were ALL blown out.)
JRA - I haven't slammed you once. I don't know why you think I have, or why I would want to.
The fact of the matter is that we have a complete failure to communicate.
If we could meet in person, we could probably straighten it all out over a cup of coffee. But with typed words? It's not going to happen.
That being the case, I'm going to stop responding to your posts after I finish typing this one. I think that will increase the amount of happiness in the world, which is always a good outcome.
One of the things that I like about engineering, math, and science, is that these fields have an intrinsic honesty to them. That honesty is important to me, and to the continued future of civilization. They are the only fields in which our fallible human species is able to be honest. That's because honesty is forced upon us by the scientific method. A politician can BS all day and not get caught in a lie. A mathematician who says "1/3 is greater than 1/2" can instantly be proved wrong by anyone with a smattering of mathematics.
Because honesty is vital to all science and all technology, I won't blindly agree with every comment made on this forum. If I know a statement is incorrect, I will disagree with it. Not rudely, not with any intention to "slam" anyone, but simply because it is a factual mistake. The disagreement is not about the person - it's about the information content.
The fact that I don't agree with everything that's posted can be misunderstood in this era of mindless political correctness, in which we are supposed to treat every opinion as equally valid. People think it's rude not to agree with everything. I think civilization will collapse if we agree with everything.
I didn't (and don't) agree with some of the technical content of some of your posts. Maybe that's why you think I'm attacking you. I'm not. I'm simply disagreeing with the facts, as a matter of honesty, as a matter of ethics, as a matter of scientific integrity.
Goodbye, and good luck
-Gnobuddy
+ 1000
Opinion? Be my guest ....... (but then also listen to mine)
Numbers? Scientific fact?
Arguing evidence is as useless as shouting at the clouds.
The P27 requires either something in the FX loop (or a preamp like the one suggested), replacing the diodes with a FuzzFace circuit or something, or better, replacing the diodes with one 12AX7 each half clipping one half the waveform. Adding a directly coupled cathode follower after that (also a 12AX7) should blow everyone away. Down and dirty would be a Nixie tube power supply but it probably needs filtering and won't sound as good, or have the compression provided by higher voltages (like 380) and massive grid stopper resistors. This is not a fun job, but it will sound similar to a Marshall. I extensively researched the solid state power section question and I found there is no point in using anything other than the P27 for a variety of reasons. It was designed specifically for guitar and handles a 4 Ohm load....with the usual proviso that it will sound like every other high-NFB solid-state guitar amp: clean tones will be too clean, sounding thin and cold when you play an electric guitar through it.
My experience is that this sort of Hi-Fi clean preamp works well for electro-acoustic guitars, acceptably for hollow-body archtop electric guitars in a jazz context, tolerably for semi-hollow electric guitars (like the ES-335), and very poorly for solid-body electric guitars ('Strats, Les Pauls, etc).
Too-clean is not the only problem with this type of SS guitar preamp, built around op-amps and textbook Hi-Fi building blocks. Elliot claims his low-impedance diode clipper (D2/D4/R14, here: https://sound-au.com/project215-p27-revisit.htm) will distort less harshly than more common higher impedance versions of similar circuits. That may be true, but only a mother, ahem, Hi-Fi audio engineer, could love the raw sound of an electric guitar fed through two reverse-parallel clipping diodes.
Physicist John Murphy figured out, decades ago, that for overdriven guitar to sound good, the duty cycle of the output waveform needs to vary with signal amplitude. This happens automatically with vintage capacitor-coupled guitar amplifier designs, as average bias current through the various gain stages unintentionally shifts with input signal level, due to the nonlinearities in the amplifying device.
However, duty cycle modulation doesn't occur at all with textbook op-amp gain stages, or the sort of anti-parallel clipping diode approach Elliot uses in his preamps. Instead you get a predictable and perfectly symmetric output waveform, which sounds like a predictable and boring buzz.
John Murphy designed an op-amp circuit that created duty cycle modulation when overdriven, by using a small-signal diode to deliberately shift the op-amp operating point dynamically with signal level. That circuit was used in a number of Carving solid-state guitar amps over the years.
I found sound clips of some of those old SS Carvin amps on the 'Web a few years ago. To my ears, guitar clean tones were thin, sterile,cold, and unpleasant, just like every other op-amp based electric guitar preamp. But overdriven tones were definitely richer and better sounding than the generic diode-clipper circuits we've all been listening to for years. Murphy was definitely on to something.
I think a combination of a JFET stage (or stages) to add a little low-order harmonic distortion to warm up the cold clean tones, and John Murphy's duty-cycle-modulating op-amp distortion stages, might be very interesting to try. Perhaps that will finally produce an analogue SS guitar preamp that actually sounds halfway decent.
Then again, you could simply plunk down $76 (USD) and buy the remarkably good-sounding Flamma Preamp discussed above.
This old (1994) article about John Murphy makes interesting reading: https://www.trueaudio.com/at_eetjlm.htm
-Gnobuddy
I also researched if the dynamics of tube circuits could be replaced with solid state like the Carvin attempt, and found that if it were feasible, tubes would no longer exist in guitar amps. This is my opinion, but I will always support experimentation and accept the subjective opinion of others.
Umm, what is a "P27" ?
Purple twenty seven? Pussycat twenty seven? Pneumatic twenty seven?I rather like "pussycat twenty seven" myself. In my mind's eye, I see a green racetrack with highly-strung animals waiting eagerly to burst through the gates and race to the finish. Only the animals are all pussycats, not horses.
Pussycat Twenty Seven looks particularly handsome, with a shining coat and an excitedly waving tail held aloft. She's raring to race, unless she happens to spot a butterfly or see a grasshopper along the way.
I think some of the old Carvins with John Murphy's "duty cycle modulation" op-amp overdrive circuit produced interesting sounds (I found some sound clips online a while ago). The overdriven tones from these are certainly not as rich as a good tube amp, for sure. But also not as boring, buzzy, and nasty as ye olde diode clippers that everyone else is/was using.
You may very well be right, though I think KMG came very, very close with his guitar amps using his MOSFET tube emulation stages. And whether or not they sounded exactly like a tube amp, they certainly sounded like very good guitar amplifiers. I think the solid-state guitar amplifier industry missed a big opportunity by not licensing KMGs work and using it to mass-produce guitar amplifiers.
But let's say you're right, and solid-state analogue will never sound as good as tubes. What then?
Here's the simple truth of it: vacuum tubes have already outlived their lifespan by at least sixty years. They were supposed to have disappeared by the time the 1970s rolled around. The fact that there are still new vacuum tubes being made in 2022 is amazing, but it is also obvious that the tube manufacturing industry has "one foot in the grave and the other on a banana peel", as one of my friends used to say.
Earlier this war Putin's bloody invasion of Ukraine not only cost tens of thousands of lives and caused untold human misery - one small side effect was that it also precipitated a massive vacuum tube shortage, along with enormous price hikes. For a while, JJ Electronic in Slovakia was the only factory in the entire world still making and shipping new tubes suitable for electric guitar amplifiers.
Between high prices, limited availability, and fragile supply lines, the end of the vacuum tube guitar amplifier is already here, in the same way that the end of the horse-drawn wagon is here. Just as there are still horse-drawn wagons in existence, there are also still tube amps out there, and there will be for years to come, in ever-dwindling numbers, mostly living out their lives as expensive luxuries in climate-controlled recording studios.
But the era when every gigging guitarist lugged around a tube guitar amp is already over. The era when beginner guitarists started out with a Fender Champ tube amplifier is over.
A lot of pro guitarists, even high-profile ones with A-list clients, have already moved to Axe Fx and similar solid-state tube emulators. Working musicians (the ones who survived the COVID lockdowns) are switching in droves to less expensive digital modelling solid-state amplifiers like the Boss Katana series, and Fender's new Tone Master series.
The future of electric guitar amplifiers is already clear. The cheapest and nastiest ones still use all-analogue electronics, with op-amps in the preamplifier section, and IC power amps. They sound just as bad as they did decades ago, but they still sell, so they are still made.
Beyond that, DSP chips running digital models emulating tube-like behaviour are taking over the market, over almost the entire price range.
But what about DIY? What about hobby builders? That's the one remaining tiny little niche where solid-state, analogue, guitar amplifiers still have any reason to exist.
-Gnobuddy
Elliot Sound Products (ESP) Project 27A Solid State Guitar Amplifier. PCB with B/M and instructions available. This eliminates the weight and the expense of the OT, the unreliable bottles and tube rectifier which are the first to go, or worse, intermittent failure. Preamp tubes can last over 50 years. No attenuators or bottle-pulling for practice level required. No impedance matching taps on OT required/desired. No need/desire for Depth and Presence controls. There is a loss of power tube clipping, but for me, I don't see the loss of this feature enough of a deficit to justify the shortcomings of the original Fender/Marshall designs. I am not against those old designs, but the filled a need in an era that no longer exists. And older folk don't want to lug all that weight around anymore, or sending it to a tech every 5 years when it acts up, if they can't DIY it.Umm, what is a "P27" ?
****
But let's say you're right, and solid-state analogue will never sound as good as tubes. What then?
****
But the era when every gigging guitarist lugged around a tube guitar amp is already over. The era when beginner guitarists started out with a Fender Champ tube amplifier is over.
****
-Gnobuddy
I'm sure tube simulator amps sound very good, but what happens when they stop working? Even a DIYer would struggle without a diagnostic manual, if they even exist, and after 10 years no replacement parts/PCBs !! Tube amp acting up? Swap tubes, check for bad solder joints, loose dirty sockets, worn out pots, dried out electrolytics. 99.9% of the time the problem is found in one of those.
True, a beginner does not need a tube amp to learn, but he/she will probably get one once they do..... I learned with a Pignose and the MXR Distortion Plus. It sounded very good and exposed me to sensitivity and compression. I saw some very small, low voltage, wafer-sized "tubes" somewhere ( a while back) but they were expensive and no reports on how--or if--they clipped properly for distortion. I should look to see what the status of these are. If the HT/heater supply can be eliminated, that would be great.
The op-amp is definitely the dominant player in solid state guitar preamps and in a way the IC power amps are much like the op-amp, cheap and easy to implement one chip solution. We've talked about discrete designs using Jfets and Jfet/BJT pairs as "gain blocks". In a way it's too bad that no one (that I am aware of) designed a purpose built integrated version IC for musical amps that wouldn't clip abruptly, wouldn't be perfectly linear etc.. and not necessarily based on operational amplifier topology. I like working with discrete stages, but for mass production it was the op amp that quickly took over. Perhaps something like a plug-in mini board that would replace a common dual op amp chip. These do exist in the market as "discrete op-amps" ..but none that I know of were designed to purposely introduce low order harmonics, nonlinear response and "soft clip" limiting.