Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - J M Fahey

Looks VERY good in general.
Specially the third screen capture which shows a waveform with a kink near zero crossing plus rounded tops, quite close to a Pentode push pull clipping, the expected "tube power amp clipping". 👍🏻
No power transistors in parallel here, you only have single Q8 on positive rail and Q11 on the negative one, so nothing to match.

Now in higher power amps where you have multiple transistors in parallel, as in Gibson G105 (or something like that) being discussed in another thread, where they use parallel 2N6254, yes, there you need it.

That said, modern transistors are *so* consistent that if same brand and type and from same batch (say you order 4 or 8 together from Mouser - Digikey - etc.) they *will* be very close, by default.

I often buy power transistors in bulk, say 20-25 units in a "stick".
Used to measure all and attach small labels to each ... not any more, incredibly consistent.

Only justified if, say, one transistor is made by, say, Fairchild and the other by ST or ON, or one was bought in 2014 and another in 2023 or some similar separated vintage.

Even so, low value emitter/ballast resistors (0.1  0.22  0.33 ohm) tend to even out current very well.
1) agree that in SS amps, supply voltages often match or at least are real close if power is about same (here 65 W) and supposing speaker impedance is same (8 ohm?.so power transformers should be more or less interchangeable.
Only problems come, sometimes, if main winding voltages match but some get preamp voltage out of main rails, by attenuating those using Zeners or regulators (as in Fender, Laney, Marshall, etc.), while Asian ones add extra 2 x 15-20 VAC secondaries.
So if you need those you _might_ have to buy an extra small transformer for that.
Often happens with popular Antek toroids.
Amplifier Discussion / Re: Gallien Krueger 800RB
September 02, 2019, 09:10:08 PM
What G1 said

Failure description + pictures point at gross overpowering.
That Bass Player must LOVE Jack Bruce sound.

- "hey, your Bass is distorting ..."
-" what distortion? ...."

Full rail to rail squarewave all the time.

Probably 500/600 W squarewave from a 300/350W RMS sinewave power.

4 x 10" voicecoils (150W each) stood onslaught for 90 minutes; single EV one (probably oldstyle EVM15L rated 200/250W RMS) stood it "minutes".

I see a failure which wass more commong long ago, not that much today: even before wire to wire glue toasting, whole flatwire block unglued from Kapton former.

Actually epoxy attaches *poorly* to Kapton, too smooth no pores surface and this failure was relatively common; afterwards some mechanically roughened winding area with a process similar to sandblasting (I used fine grain wetordry sanding paper) or chemically etched surface, dramatically improving adherence.

I bet that EV speaker can be reconed with modern aftermarket coils and stand more.

In any case, Bass player is some kind of Terminator, since he blew the 4 x 10" too.

Given his style he needs 2 x 8 ohm 4x10" cabinets  with 8 x 10" 150-200W RMS speakers ... since he is applying constant 500/600W (clipped)  to them.

EDIT: almost forgot.

before moving so much that it actually clipped the wire opening the circuit, that voice coil jumped all over the place, and winding scratched gap walls both scratching enamel which already makes UGLY mechanical nouise, but also allowing wire to wire arcs appear, which spark and produce buzzy sound.

Sound must have been *horrible* but clearly player doesn´t care ... or even notice it.  ::)

Not sure amp is at fault, and quite certain user woild cause same damage with an equivalent power amp, such as a Hartke 3500 or some Ampeg SVT Pro. (the SS ones).

Or one of the small Class D ones, some of which are deadly (say 550W RMS into 4 ohm).
OK, let´s separate power supply and amplifier and measure.

With a sharp boxcutter cut tracks leading from +V and -V to amplifier at points 01 and 02.
Cut a narrow 1mm strip or so and remove copper , later you can scratch soldering mask on both sides and rejoin with a drop of solder or a little bit of wire.

1) Measure +V and -V to ground at points 04 and 06 .
If normal (35/40V ) , also measure +/- 15V rails at ungrounded ends of ZD1 and ZD2 .

2) with amp OFF measure resistance to ground from points 03 and 05.
Both ways and using resistance (say 200 ohm or 2k) and diode scales.
Open circuit is no problem, I´m trying to find some continuity to ground which shorts raill voltage.

I also suspected some open track or connection, but you confirmed
QuoteThere is continuity back to -V on C61 and DB1
so that seems to be working.
It´s designed to grab attention  :cheesy:

Now that he actually sells it, is completely different; price is outrageous, it could command $20/25 TOPS if it were in good working order, which it is not.
I have never seen such RUSTY jacks and handle, not sure it even works and certainly any control there must be scratchier than somebody rolling naked in Poison Ivy.

    Paralleling *voltage*  sources is a disaster waiting to happen.
    The additional condition that now gain must track each other within 1% or less, any volume, any frequency, plus NEEDING cheesy series resistors to somehow help speaks volumes about what a poor Engineering choice it is.
Well... maybe. Pretty much no really high power amplifiers in existence get by without emitter resistors. The emitter resistors are there to wash out the microscopic differences in Vbe.
Please don´t compare pears to oranges.

1) Transistors in parallel are emitter followers, either real or quasi complementary, with a gain=1 so a small value resistor is *ample*  to equalize current.

And if you have a Vbe difference, say 50mV between two of them (which is GROSS), at full power driving some load, if one puts out 5A through 0.33 ohms so dropping 1.65V, the other, due to the 50mV difference (say it´s lower, but analysis is the same if it´s higher) it will apply 1.6V across its own 0.33V , so current now will be 4.86A , a 3% current difference, which is NOTHING.

* And in any case current from both is "useful" since it flows into the load.

* And emitter resistors "disappear" because they are inside the NFB loop.

* And since there are not "2 brains trying to control the same muscle and opposing each other", even 5% or 10% resistors everywhere are perfectly acceptable.

2) Now to parallel voltage source, high NFB amps.
To begin with, *each*  will try to keep output voltage tightly controlled, and obey its own offset and NFB; will NOT cooperate with the other amp but fight it, feeding as much current as needed into the output node as necessary to achieve that.
Fine if driving a passive load , such as a speaker or resistor, a disaster if feeding a node which is also fed by another voltage source, worse if the other one has same NFB and same muscle because current will flow from one into the other trying to keep shared output node at what HE thinks it should be, and even mV difference can mean AMPERES of current .

Which are WASTED Amperes flowing from one to the other, instead of into/through a load.

That´s why I accept *bridged* amps (load is connected *separating*  them and current has only one path: through it) buit hate parallel ones like the kludgy mess they are.

TYrimming NFB within 1% (or else) and adding cheesy 0.1 ohm resistirs *outside*  NFB only confirm that is a poor design.

Why do they exist then?
2 main reasons:

a) not too qualified "weekend designers" find they can "design high power amplifiers::) without all the nasty Math, just pile up LEGO building blocks and pray.

b) you will need to buy MANY chipamps to build them, and also allow typically low to mid powered chipamps occupy a market sector they are not really qualified to= $$$$$$ for Chipamp makers  ;)

Now as I said before, if you use them *separate* , each one feeding its own speaker , fine with me.
Now that I think about it, you will probably have more Preamp signal out as well.
That cheesy power amp has very low input impedance (think about 1 k or so), and loads the volume control down a lot; I think when you remove the old one and add the TDA2003 instead, which has higher input impedance ( 20k to 40 k or so) you will now have more signal available at the signal out jack.

In fact I guess you´ll need to add a second volume control, so one for the internal amp, one for line out, OR regulate just the internal amp, and send line signal "straight" .... which in any case will be regulated at the external amp.
Quote from: R.G. on June 17, 2018, 01:31:56 PM
As usual, I have a mildly heretical viewpoint.   :)
Change that label to *creative*  and I´ll buy it :)

QuoteChip amps do have fundamental limitations on their ability to dissipate power. As a practical matter, the TO-220 packages like the LM1875 can only really do a 20W amp out at the edge of reliability. The LM3886 and similar packages can get to about twice that and be reliable without gigantic heat sinks.
Just today I posted about the same on some Forum.

LM3886 have an impressive realiability record ... TDA729x not so much (and that´s an understatement).

Why?: both use the exact same case, one is expected to deliver 50W .... reaching 60W only at *reduced* +V rails, the other is happily touted as a "100W amp" and proudly says so on datasheet, also is happily touted as using +/-42V rails in Real World applications (as opposed to Lab tests) ... we all know results, Marshall had to make plug in modules so users can swap TDA729x themselves  :duh
My solution to this is to use the LM3886, which is an incredible deal at about US$5 each, and use more of them. There are circuits to parallel them up for higher currents and better power dissipation, and to run them bridged for higher voltage. However, I would not use them that way.
Fully agree: bridging is a nice way to turn 8 ohm loads into 4 ohm ones, and 4 ohm into 2 .... a mess and most nobody notices that.

Paralleling *voltage*  sources is a disaster waiting to happen.
The additional condition that now gain must track each other within 1% or less, any volume, any frequency, plus NEEDING cheesy series resistors to somehow help speaks volumes about what a poor Engineering choice it is.

Ultimate madness of course is bridging parallel TDAs , somebody at DIY Audio even proposed triple parallel bridged ones.  :loco :duh

Worst thing is that by then the amp, besides being way less reliable, has become more complex than a properly made discrete one !!!!  ::)

QuoteThe LM3886 is quite reliable (given a decent heatsink) at 30-40W output. Most guitar amps run one 12" speaker on about 30-50W of amplifier power. When it gets over that most amps run two 12s or four 12s. Why fight the 30W per speaker practice? Why not make one highish quality amplifier per speaker. They're easier and more reliable that way, as well as more adaptable to situations and portability.
Excellent idea.

QuoteI just went through this in designing a replacement/repair power amp for the Thomas Vox amps. I opted for using an LM3886 per 30W speaker load. This covers the Buckingham and Viscount with 8 ohm loads with one amp each. The Royal Guardsman needs two modules to run two 12s at 60W, and the Beatle needs four for 120W.
Not sure I follow you: you use 4 x 3886 in parallel to drive a 2 ohm cabinet through a single cable pair or a special connector with 4 individual speaker cables?
Not problem in a combo, not so sure in a head+cabinet situation.
QuoteIt is far easier to get a 30W amp to run right than to get a 120W or higher to run right. And replicating several small things can be simpler than making one big one.

I did a PCB that is about 2" by 3" and runs one LM3886. I found a suitable heat sink for under $8 at Antek offers the AS-0522 toroidal transformer for $17.50, and this is a nearly ideal transformer to power the LM3886. You can fit the whole amplifier on one heatsink serving as a sub-chassis, power supply included.

And if one fails, you've thoughfully made one more power amp module than you really needed, so you swap in the spare while you fix the broken one.

Just sayin' ...
Please check that Tr10 collector is insulated from heatsink and chassis.
Did you use mica+grease or a Silpad there?
Collector shorted to ground would kill -V .

Or: check for continuity between Tr10 collector and -V terminal at main filter caps (C61) and/or - end of rectifier bridge (Br102/DB1).

Or maybe negative-pointing part of bridge (Br102/DB1) is open.
Thanks Phatt.
Yes, since last October I have been very busy, revamping the new amp line, designing new models, etc.
Not much free time to visit my beloved SS Guitar Forum.  :-[

here´s some of the new stuff, which I introduced at the 2018 Convention of the Audio Engineering Society Buenos Aires branch:

Amplifier Discussion / Re: Sunn Coliseum 300 Issue
June 17, 2018, 03:52:53 AM
problem isn that meter does not read "just"  the resistor but also quite a few stacked resistors which are in series: 2 probes , 2 wires, 2 banana plugs, 2 or more rotary switch contacts inside the meter; all easily add up to 0.4 to 1.5 ohm, specially when corrosion starts to appear, simply because of damp air.

Normal is to short probes to each other, it will NOT show Zero, note that value and then substract it from resistor measurement.
So if you have , say, 0.6 ohm "parasitic" and resistor measures "1 ohm" , it actually measures 1-0.6=0.4 ohm .

Also notice that at best you get the first digit right, say 0.3 or 0.4 or whatever; FORGET distinguishing 0.33 from 0.39 for example. 

So, how do factories check production?

There are other ways, simply too expensive for the average Hobbyist or even regular Tech:
* impedance bridge, this is the one I use, can easily measure 0.005 ohms

or a "4 wire meter" which uses 2 wires to feed current to the resistor and 2 separate ones to measure.
Since there is no current passing through the "measuring only" ones, error is avoided.

But for simple "Go/No Go" everyday testing, normal meters and error substraction are enough.
Excellent suggestions by phatt, as always.
I´d like to add about:
QuoteI know they are simple-but-tough little critters and you can push them to get better power,
Sadly, not any more.
They were discontinued 2 years ago, last 5000 were *swept* from Mouser shelves in a couple hours, I got 50 myself ... and that was it.

As many as you want from China/Hong Kong, pity they are filled with wet sawdust or worse.

Tayda can be trusted, and if they are out of a certain component, it´s because they can´t get the good ones.

LM1875 is still available, but it´s 20W tops .
In any case I prefer real 18/20W to unreliable 13W provided by fakes.

TDA7294 ia a good chipamp but Marketing Department advertised them as "100W amps"  whichb they are not, and die trying to comply.
Use them for 70W RMS or even better 50W RMS (which means rails about +/-30V tops or "whatever LM3886 uses") and they will work forever.

Whatever LM3886 puts out (50/8 or 60/4) is what can be dissipated by that tiny case, which amounts to two TO220 packages side by side, you can´t go against Physics Law and pretend it can safely dissipate more heat because it can not.

So just make that simple power amp and then experiment at leisure with Preamps; results can be very good.

And real 50/60W into a good 12" Guitar speaker is all it takes to play in a live Club situation, alongside a drummer and other Musicians.
Amplifier Discussion / Re: Sunn Coliseum 300 Issue
June 16, 2018, 08:20:43 PM
Just one detail: low value (0.33 ohm and such) resistors do not "go out of spec", they are either fine or burnt open.
But regular meters have big trouble reading such low values.

So if a 0.33 ohm one shows , say, 0.6 or 0.8 ohm ... trust the printed value rather than the meter  :o

meaning: don´t replace them unless actually open.

Worst case is that you may replace an "out of spec "  one ... and later find "the new one is also out of spec" ... which it is not.
Just in case we are missing something , let´s recheck voltage on all pins.
It doesn´t hurt and we might get a surprise.
So check and post voltage on *all*  pins, even those which "officially" are not connected.

Main problem on the old one was 0.22V on Pin 12, which is speaker out, and should be around +9V, but something I found somewhat weird was you found full 18V on Pin 4 ... it´s a bootstrapped output, is fed by a 100 ohm resistor and some current passes through it, so thanks to the series 100 ohm resistor it should be at least 1V below 18V ... yet you reported 18V with no losses.

Of course, since chipamp was damaged, that was possible.

But with a new presumed good amp it should show what I expect.
In fact in my original post I suggested "3/4 of 18V".

So remeasure and repost, but feel confident that absolute worst case we can Mod this amp to use an easy to find TDA2030 or LM1875.

But let´s try once more to make the original work.  :tu:

Almost forgot: use a loupe or strong looking glasses (+3 or +4) and look around all chipamp solder pads and tracks leading to them ... we might have a broken/cracked track.