"Black Flag" JTM100 - the build has begun!

[shakti]

Done! First I got the filter caps formed nicely down to about 1.70V at the lowest. I had to use a 27k in series with the bias diode to get the bias voltage within decent range. Tubes were redplating at first, and bias fluctuating quite a bit, even with the 27k resistor. Eventually the tubes settled in decently, I got sound on first fire-up, sounded veeery nice and quiet, noise-wise.

I only got to try it for a few minutes severely attenuated (Alex' Attenuator). I do have one problem, where I get some pretty severe crackling, or whatever I should describe it as, when I really dig in. If I whack the low E, there's a loud crackle on first impact, and a few tiny bursts of crackle immediately after. Could it be a blocking distortion of some kind? I've never encountered it before. It's too late to do much trouble-shooting. Preamp tubes are known good ones, power tubes a less well known entity. All the voltages checked out fine, except I don't get enough range on the bias pot, had to turn it all the way to get the bias low enough.

Any hints are appreciated! I'm gonna check all the solder joints, but I'm pretty sure they're all good.

Apart from that nasty crackle/distortion, my immediate impression was WOW! The sustain was beyond belief! It sounded like my guitar was hardwired to the speakers! Easily the best sustain and feel of all the amps I've built...now, just to get that crackle sorted out...

[shakti]

Well I just love problems that solve themselves...this morning when I plugged in again, the nasty distortion was completely gone. No idea what it was, maybe I should check the speaker cable...

I've been busy all day, so no photo and only a few minutes more playing time. I'm really digging it - it's still a bit buzzy and fizzy, and the bass farts out a bit at high volumes, but I expect that to improve a lot with burn-in. What I do absolutely love is the way it compresses - it stays very firm until around 7, then smooths out and compresses a lot at the end. The legato sound at 10 is stunning! This amp really does "Crossroads" tones to a T. The clean-up via the guitar is also tremendous, probably better than any of the other amps I've built in that respect. Maybe it's the dual rectifier?

[Roe]

Well I just love problems that solve themselves...this morning when I plugged in again, the nasty distortion was completely gone. No idea what it was, maybe I should check the speaker cable...

I've been busy all day, so no photo and only a few minutes more playing time. I'm really digging it - it's still a bit buzzy and fizzy, and the bass farts out a bit at high volumes, but I expect that to improve a lot with burn-in. What I do absolutely love is the way it compresses - it stays very firm until around 7, then smooths out and compresses a lot at the end. The legato sound at 10 is stunning! This amp really does "Crossroads" tones to a T. The clean-up via the guitar is also tremendous, probably better than any of the other amps I've built in that respect. Maybe it's the dual rectifier?

damn, I may have to build one of these. A couple of questions: did the first dual rectifier amps come with 270K mixer resistors and the last ones with 470Ks? and did all use the 27k nfb (while late 67 or early 68 used 47k)?

my 67 improved a lot with break in and some tweaking. it went from good to great but it took quite a few months.

btw: got the alex att. seems to work fine although the on/off switch is contra-intuitive

[shakti]

Cheers on the Alex! Yes, the standby switch is a complete mystery...but Alex likes it that way.

The early black flags did indeed have 270k mixers, and the later ones 470k. Same thing with the JTM50s. I built mine with 270k, as I was shooting for the early-mid 67 circuit that Hendrix most likely had in his black flag.

I was going to ask about the NFB. I've never seen anything but 27k in black flag amps. 47k would have been very, very late in '67 at the earliest. But I'm unsure about the output tap - I put mine at 8 ohm, but I wonder if the early ones could have had 27k@16 ohm?

Neil - what does your block end chassis EL34 amp have? 8 ohm or 16 ohm?

[Roe]

Cheers on the Alex! Yes, the standby switch is a complete mystery...but Alex likes it that way.

The early black flags did indeed have 270k mixers, and the later ones 470k. Same thing with the JTM50s. I built mine with 270k, as I was shooting for the early-mid 67 circuit that Hendrix most likely had in his black flag.

I was going to ask about the NFB. I've never seen anything but 27k in black flag amps. 47k would have been very, very late in '67 at the earliest. But I'm unsure about the output tap - I put mine at 8 ohm, but I wonder if the early ones could have had 27k@16 ohm?

Neil - what does your block end chassis EL34 amp have?

the early el34 amps I've seen pics of, the super 100 amps, have all 27k. typically at 8 ohms. although larry has seen a 67 with 27k at the speaker jack

[Roe]

It appears greg g went from the dual rectifier circuit to the later single rectifier (and upped the mains to 50uf):

"The dual rectifier bridge is replaced with one very similar to what you see in an early 12,000 series head. This is period correct for the latter part of 67. Initial filtering is increased while the rest of the stages stay filtered very low. A 5 Henry choke is used Drake style 352-114. I don't feel the need to go with a larger choke. Sporadically during this time period you will see a R/S 10H choke used but not that often. The larger choke induces a greater voltage drop across the screen supply."

http://www.vintageamps.net/plexiboard/v ... 66&start=0" onclick="window.open(this.href);return false;

[Flames1950]

Somebody drug up my old thread at Plexi Palace eh? The pics are gone though, deleted them ages ago. I still have that Black Flag head, since I loaned it to a forum member for a couple of weeks and UPS busted it up in return shipping -- bent the chassis and the transformer mounts. I like it better with 50uF on the mains too, with the rest of the filtering stock.

Still had a pic of the second one I built hosted though, before it got butchered by a later owner:

[Doug H]

maybe a problem in the b+ rail if you notice it most on the low string hit hard, ie, big power draw equals big problem?

[Roe]

I also prefer 50uf on the mains on my el34 super 100 amp. otherwise the filtering is stock

[shakti]

As mentioned the problem has disappeared. Maybe it was also a "breaking in" kind of issue, maybe with the signal caps? Whatever it was, it's gone, and I'm really digging the sound. I'll upload photos later today, first I'm going to play it a bit and check voltages again, play around with bias settings.

I might just have to build a late '67 circuit at a later point just for kicks...i.e. 470k mixers, single rectifier, 50uF mains. If I were to do that, I'd probably try a 1202-132 for variety.

[shakti]

Here is the finished amp in all its glory. Sorry for the poor quality photos.





With 236VAC in I get 463V on the plates with the 240V tap, vs. 509V with the 230V tap. I tend to prefer EL34s with a slightly lower voltage, so I think I'll go with the 240V tap. Wall voltage tends to hover around 235V. One older schematic for these amps shows 470VDC on the plates. At the higher voltage, the tendency towards fizz and sizzle was even more pronounced, but I only have 1-2 hours on it yet, so I fully expect things to smooth out.

I have a problem with one of the Tesla EL34s (late production which I got from Tube Tramp) which has a tendency to redplating and unstable bias. If I bring the bias up slowly it's fine, but when flipping the standby switch it seems that tube tends to draw more current and go into redplating, even at the same bias voltage which works fine when bringing it up slowly. I'm gonna try some RFTs to see how they fare.

Some things I've noticed:
- this amp isn't the smoothest, to put it mildly. It's a pretty raucous beast. It's strange in that it seems to have a very fast attack, but then immediate compression and a perceived sag on the sustain of notes. However, sustained notes can have a tendency to some buzzing transients. I'm speculating that the very efficient rectification, coupled with ineffecient smoothing, could have something to do with this.
- at full tilt, it compresses a lot on the sustain as mentioned, but the somewhat buzzy transients makes it sound both fluid and very nasty at the same time. I guess that's what Greg Germino meant when he said it could be "unforgiving on single note runs". This "sizzle" and the fast initial attack does lend it a little bit of Hiwatt character, as others have noted.
- ghosting isn't nearly as bad as I feared. Sure, it ghosts at 10 on a neck pickup with tone rolled off, but not really much more than my JMP50 or JTM45/100, and the fluid sound I get at that setting more than makes up for it. Still, would be interesting to try RIFAs or Sozos, if they ever come out with 32uFs.

[shakti]

Hmm, I tried the RFTs, and I get the same problem: adjusting the bias from cold up to a "normal" bias range of about 65-70% dissipation causes no problems. Bias stays rock steady, and this is a nicely matched quad. As soon as I switch to standby, and then switch it back on again, the tubes all get problems "finding" their bias, and it shifts up and down, and I see a "pulsing" sort of flashing from deep inside the tubes (which I interpreted as redplating). With the bias oscillating up and down the plate voltage also swings. What on earth causes this problem? Does the switching of the B+ (via the standby switch) cause some sort of noise that oscillates and affects the bias in some way? I'm stumped...

[myfoot]

Bad/loose connection on the tag strip by the fuses? Looks great BTW. Hopefully silly little problem you're overlooking at the moment. Good luck.

[Doug H]

cool, your resistor slection is the way I'm going to do my 45/100, with the board spanning ones CC and the rest cf. I've sseen a lot of old photos like that.

[shakti]

Bad/loose connection on the tag strip by the fuses? Looks great BTW. Hopefully silly little problem you're overlooking at the moment. Good luck.

Nah, that's just a makeshift voltage tap selector, inspired by the "selector" in my '65 AC30. It either works or doesn't, so that's not it. Thanks anyway.

I just plugged in again after having it switched off a while, and now there's a little more transient noises and artifacts, and I could swear I had the output volume veeeery faintly fluctuating up and down fast, like a tremolo. I don't understand this...bias voltage at pin 5 stays steady (I have to check again if it stays steady after switching the standby, but I'm pretty sure I already checked that), but measured bias current (via 1 ohm resistors to ground) oscillates...