Quote from: CraftyZAThis Alemic circuit

Ah.  That actually says 300V not 350V which is why I was wondering.


Points arising from f2b.png;

- as the loaded supply is around 300V then the caps that are connected to this rail all need to be rated at 350V or better if they are to survive the peak HT voltage before the valves warm up.

- the anode loads R4 and R8 do not need to be anything like 2 watt rating, half-watt is more than sufficient.

- the tonestack is a classic Fender with identical values, so the mid control should be 10k linear not 250k as shown.  250k would pretty much disable the treble and bass controls.

Quote from: CraftyZAsupplier gave little tiny ceramic caps for everything

The less you specifiy, the more they will give you the cheapest thing to hand.  You really have to find your suppliers' specific part numbers for the caps you want (or spell it out on the order e.g. "0.1uF 100V polyester" - but it's less reliable because it depends on the parts picker knowing the difference between a poly and a poke in the eye, and they tend to mindlessly go by part numbers).

Quote from: CraftyZAThey only show the one on the b+ rail which is 350v

You need to be specific which circuit you are talking about.

It's a fair bet that the others will also need to be the same voltage rating because they will rise to the unloaded power supply voltage before the valves warm up, draw current, and pull the voltages down a bit.

{An Irishman and a large crocodile walk into a bar...}


Hi ogeecheeman, welcome.

Quote from: ogeecheemanuser replaced bridge rectifier.installed backwards.

Bless their little hearts.

Quote from: ogeecheemanrail voltage on test lamp...+34vdc / -34vdc.  collector and emitter voltage on ALL output transistors = -21vdc.

This is contradictory, so you need to find why you are measuring +/-34V (presumably in the power supply) and -21V on both the output stage supply rails ("ALL collectors").  This will be an off-circuit problem such as blown tracks.

via's or plated-through holes from one side of the PCB to the other have been known to blow out like fuses.  Start with your + and -34V rails and trace them through towards the OP collectors.

Quote from: galaxiexInspiring/Philosophy/History/Technical/Etymology/Trivia/Math/Diagnosing/Design/Culture

"Everything is deeply intertwingled.  In an important sense there are no 'subjects' at all; there is only all knowledge, since the cross-connections among the myriad topics of this world simply cannot be divided up neatly." - Computer Lib/Dream Machine, 1974, Theodore Nelson.

Google seems to treat "circuit" and "schematic" as synonyms.

I don't think "reasonable" is a word that could be applied to a 160 watt preamp.   

Ah yes, life.  I seem to remember having one of those once.  {spent last weekend under SWMBO's car, restoring some domestic equilibrium.  I'm getting too old for this malarkey.}

Quote from: cbg RickI'm sorry but, you keep referring to this amp as having Ge transistors, I checked the transistors in the parts list and I'm not seeing any Ge transistors listed. Am I missing something?

No, you may well be right, but as teemuk points out above, this is the least of your redesign problems.  This is still a case of "jack up the hubcaps and put a new car in-between".


The V_BE multiplier is only one way of getting a temperature sensitive bias, such as a string of diodes, but only the transistor circuit operates as a "base-emitter voltage multiplier" (thanks to the gain of the transistor).

Quote from: cbg RickWould it also work to use the BE junction of a TO-220 transistor to replace the diodes and clamp that transistor/diode to the same heatsink? 

It not only works, it's my personal favorite.  BD139/140's for example have a very handy mounting hole through the middle.

Quote from: cbg Rickso I am going to forge ahead. :cheesy:

Popcorn?  Check.
Hard hat?  Check.
Smoke mask?  Check.

Quote from: JLTHere's the link to that product:

...and not a word about the need for high input impedance with piezos, all the more so with smallish ones like this.

I'd be thinking of a high impedance FET buffer as close to the piezo elements as possible, built in, beltpack, or FET lead.

This circuit would also work.

If you untangle the signal path around the "Pre Out 1/Dry" and "Pre Out 2/Mix" (middle of your posted circuit, ref: C5) you will see that the chorus is only applied to one output channel.  The internal chorus return is at C66 (ref: B4) and is switched via FET Q15.

I just had a look at the Roland Jazz Chorus circuit and they do the same thing, chorus to one output only.

Try dropping your "LDR type trem circuits" into Google image search.  {trade secret}

Yeah, it's not unreasonable to think that the light/LDR need to be matched to work, but in fact in this application light "roaches" tend to be a bit overly sensitive.

For the current source or sink a transistor or two should do, it's not like we are after precision, just a nice working range.

---

Sir Arthur Conan Doyle created the ultimately rational Holmes - "Just concentrate on the facts Watson" - but I would have misgivings entrusting my eyes or a real investigation to a man who believed fairies existed.  He was convinced by the Cottingley Fairy photographs, the famous 1917 hoax, and even wrote a book, The Coming of the Fairies (1921), on their authenticity, uncritically accepting a fraud at face value.


(the first "shopped" photo?)

When Harry Houdini's mother died they had an arrangement that he would try and contact her after death through spirit mediums, but unlike Sir Arthur, Houdini soon changed his mind and then spent a lot of time and energy exposing and debunking mediums as fakes, and this shift ended the friendship between Doyle and Houdini.

Similarly Forrest Mims is well known for his excellent work teaching electronics which I acknowledge and admire, but I have similar reservations about anybody who is a Creationist.

{I'm trying and failing to find a quote about the capacity for a person to have a fine incisive intellect in one part of their brain co-existing with utter nonsense in another part.  I've been around enough New Age Kozmic Hippies to conclude that believing palpable rubbish is not "harmless fun" but has a corrosive effect on the intellect.}

Holmes' "Just concentrate on the facts" should be writ large above every electronics workbench, but a common idea that many people have is that electronics gear has some sort of will to frustrate, or that some sort of unknowable metaphysics can operate, and is something I've been fighting most of my life.  Just because I don't understand it does not mean that there isn't well-established theory that covers it.  My tag line comes from my electronics mentor, Theo Van Bemmel who came up through the trades at Philips Eindhoven to ultimately become an electronics engineer.

In my experience in dealing with electronics of all sorts I've found that the problem isn't the gear - that just is what it is, not a devious moving target but my lack of understanding of the true nature of the fault; the problem is in me.  The human brain tends to make assumptions, snap decisions, which are then included in your analysis as "true", and I spend most of my time trying to dig these assumptions back to the conscious level and critically examine them to see if they are indeed true (double check, measure three different ways, etc). 

Being a good faultfinder/repairer is about the internal intellectual struggle to try and see the fault as it really is - a mental shift of perspective.  {ref: "Computer Power and Human Reason", Joseph Weizenbaum; The Law of the Machine, The Law of the Fault.}

I am not only interested in fixing things, I'm also interested in the mental processes that we have to go through to work out what is wrong, and how to keep those unruly mental processes, snap judgements, assumptions, under control.  Aviation procedures and pilot checklists are an example of a process that has developed to lift 60% human approximation to the 99.9% level of perfection required for safe aircraft operation.

{Checklist fail; Air Florida Flight 90, "Palm 90", waiting for take off from Washington in -4ºC and three feet of snow.
First officer (reading checklist): "Engine anti-ice?"
Captain: "Off".
Two trained and experienced brains, both asleep at the wheel.
Not their only mistake that afternoon, but possibly the most critical.}

---

Quote from: galaxiexmassive 250VA 24V-24V toroidial transformer

... and 10Meg input resistor ... you don't do things by half measures, do you?   :lmao:

The input resistor is a very minor point since anything over a Meg should be high enough (unless you are using piezo pickups).

At first eyeball of your preamp circuit it looks pretty reasonable without running a sim, but a couple of points.

The power supply is shown as two 9V batteries bypassed by a couple of electros.  These electros should be the same, and their mid point connected to the battery midpoint to ensure voltage equalisation (however...).

I've seen quite a few preamps and mixers running on a single or split 9V supply and they all lack signal headroom.  I'd use a higher supply if possible, say +/-15V.

The electrolytics in the signal path are a bit problematic given they rest with no bias voltage, and this is not really a good mode for this type of capacitor.  I'd be inclined to use large film caps (e.g. Greencaps) if possible.  {I'm not a fanatic about keeping elecros out of the signal path, it's just that they are often not really the best choice.}

Quote from: galaxiexTweaked the cap and R values and it worked! (where's the SHOCKED! emoticon?)

Ha!  You've got to do something pretty diabolical to prevent a schmitt/integrator oscillator from starting up.   

This oscillator of course is from the synth crowd who are developing synth module Control Voltages and the sharp turn-around of the ramp would be less important to them.


"Plagiarise and hybridise" - works for me.      :dbtu:

Quote from: galaxiexI just can't figure out how to drive a single LED and have a nice smooth, wide range, variable brightness control (Depth) for the LED.

... which may help to explain why almost everybody puts the Depth control on the LDR/signal side, like;


(note that this circuit depends heavily on the lag of the lamp and the LDR to soften the impulses).

Quote from: galaxiexI'm using a green LED, I don't have the specs but with my meter on diode check it measures 2.8 fwd drop.
Using green cuz I read on the interwebs that CdS cells are most sensitive to light in that range.

That's one of those things that come under the heading of "true, but not exhaustive".  While that's true it is confounded by the poorer conversion efficiency of green LED's.  I thought the same but JMF reported some of his experiments and coming to the conclusion that you may as well use a red LED (and I've learned to trust his judgement  ).  Most of these "roaches" use either a small filament lamp or a neon (valve amps), neither of which has any green content worth talking about, yet they still work just fine in a multitude of commercial amps because LDR's have a fairly wide spectral spread.




Tip; LED's are current operated devices, so what you are looking for is a voltage controlled current source or sink as your LED driver, e.g.;


The current in the LED, I₀, is equal to V_I/R_E.

Note that these LED/LDR arrangements are fairly sensitive so you should need only a few mA max, nothing like full current, and all the more so if you are using anything like a High Efficiency LED.  You want the LED to go very dim but not go right out or you may get a thump or click.

10Meg for your input resistor may be a tad over the top - I'd suggest between 1 and 2.7Meg.  (remember the higher the resistor value the more thermal noise it will produce, and this is right at the input)

Fx LFO's come in all sorts, square, triangle like yours, and sine.  Square and triangle score over sine in that they are dependably self-starting, and can be made to have a (much) wider frequency range, however the sudden step in the waveform can sometimes be troublesome as a bleed through "click", and the modulation shape is thought by some to be not as nice as sine.

You could stick a low-pass RC on the output of your LFO, with an fc about your highest LFO frequency.  That should soften it a bit.

Here is a way of doing it with a FET;

I have to say again that the redesign required to turn the original Ge+Si "Lil Tiger" into a workable Si-Si "Runcible Lil Tiger" is not inconsiderable the child will be a hardly recognisable child of the parent.  This is about as far from a kit as you could get.

As an amp designer I would, and I suggest you also, seek out a design that is closer to your desired end point to reduce the amount of modification required.

I have been searching for a suitable circuit on line and I'm a bit surprised that I can't find what I'm looking for; a bit of refinement without too much complexity.

My biases lead me to;
- Long-Tiled Pair, LTP, (diff amp) input stage, say 2x BD139
- Voltage Amplifier Stage, VAS, say a BD140
- quasi-comp output stage, say BD139+2N3055, BD140+2N3055
- simple split-rail supply.

This basic transistor selection could be improved but mainly depends on what is locally available.

Many amp designs are scalable, mainly by changing the supply rails, and may work over a wide range of supply voltages and thus powers without modification (e.g. a "50 watt" design should scale down to 15 watts).  At the same time you can scale/adjust the transistor and heatsink specs.

There is no real reason why you can't use your Flamethrower 1000 for practice (apart from the risk of dropping your guitar and having to replaster your bedroom), but there is something to keep in mind, and that is that amp specs are generally referenced to maximum power output.

If the spec sheet says "Hum and noise less than 60dB ref full output" it should be apparent that the absolute residual noise level will be quite a bit higher for a 1000 watt amp than it would be for a 10 watt amp.

In a typical on-stage situation the hum and noise of a 1000 watt amp won't be audible in the front row, say 30 feet from the amp, but sitting next to it in a quiet bedroom it will be unacceptably loud, while 60dB below 10 watts might be almost inaudible.

That said, big amp builders generally take a bit more care and try to get a better hum and noise spec, say -90dB ref full power, so that the amp residual noise is acceptable in absolute terms even in a quiet environment.  Even if it is insignificant in intended use it will be auditioned in a much quieter music shop and an audible noise level is not a good selling point.

{JMF makes the point elsewhere contrasting Fender who short in the preamp input when the instrument is unplugged, to Marshall who mute the output of the preamp to remove this source of residual noise.}

Quote from: LoudthudAC gain is not affected by the bias Voltage set pot R5 because that pot is bypassed by capacitor C3.

Yep, sorry, overlooked that.   

Quote from: JLTDefinitely not what I want for a bass amp!

Oh, I had forgotten that detail.  IMO piezo pickups are totally unsuited to bass guitar by their inherently bad low frequency characteristic, whatever the input impedance, even infinity (which we can do for you sure  ).  Are you trying to pick up a signal from some sort of acoustic bass?

1Meg is typical for a magnetic pickup, so that will do for the time being (or series 3x 1Meg).

There has to be a DC blocking cap between the buffer output and main amp input, but you only need the one.

Get whatever N-channel small signal FET's they do stock, a source follower is pretty non-critical.

The maximum power handling (W_RMS) and sensitivity (dB/Watt @ 1m) of a speaker are not really related.  A speaker may have a high power rating yet also have a high sensitivity, or any other combination.

But JMF is the real expert here - he actually builds guitar speakers.   :dbtu:

Quote from: dogfaceDoes this indicate anything perhaps on the horizon or is this just the nature of the beast ?

Transformers buzz (or hum).  They can super clamp the laminations, they can vacuum pot the windings, but the bottom line is magetostriction - the core changes size by a tiny amount with magnetisation, and like thermal expansion and contraction it's an irresistible force that can never be entirely stopped.

Try stopping next to your local electricity sub-station or distribution switch yard on a quiet night and you will hear the big trannies quietly, or not so quietly, humming to themselves.  At high load times like early evening they can make quite a racket.  In fact complaints about hum noise from sub-station transformers is a significant problem for all electricity distributors.

A tranny with a loose lamination can make quite buzzing racket, and you can often make these better by tightening the clamps or a spot of glue, but a humming tranny is a happy tranny (but a growling tranny is overloaded).

Note that output transformers in valve amps also "talk".  You can not only hear a clean sinewave (when you are running into a dummy load, naturally), you can also pick the clipping point by the onset of rasp on the clean note.  I use this all the time.


{ed: typo}

Quote from: JLTInstead I'm using two piezo pickups

In which case the input impedance will be way too low, 2.7Megs being more typical.  Will sound very thin and trebly.



N-channel FET such as an MPF102 or 2N3819 (o.n.o.).  Drain to V+, Source to ground via "RS", that 10k resistor (see, it came in handy for something!) and to the input.  2.7Meg resistor from Gate to ground, "RG".

Shouldn't need C1 and I think your amp already has a cap at C2 (input).


I'm talking voltage gain which should not be confused with power output.  15 watts into any speaker is going to make quite a racket.  Whatever maximum power a speaker is rated for it will still produced quite a noise even with only one watt.

I'll leave speaker recommendations to JMF.