183 build

[cys]

This 183 is my second amp build. The first was a TW Liverpool posted in the thread here.

I decided to make the #183 a 50W with EL34's initially. Later I plan to try it with KT66's. I also have 18W transformers standing by in case I want to move to an EL84 version.

At this point the power side is wired up, only, and I'm getting readings from 496V to 491V, B+ to B+5 without any tubes. I'm not worrying about the high readings for the moment under the assumption that once I finish wiring up everything -- and add tubes -- the voltage readings will fall into place (more or less...). I do have the preamp board populated and soldered up at this point, but I have yet to install it in the amp.

Any input is appreciated.

I've picked components that I felt I could get in several weeks. Eyelet boards are Manning style from Amplified Nation. Chassis and relay boards are also from Amplified Nation. Signal path resistors are from Welagen Amps. At this point I think I've accumulated enough "alternate" parts for another amp, almost.

I've been moving between two circuit diagrams and two layout schematics. One of the layout schematics is from this site, and one from Ceriatone. When I was wiring the preamp heaters working from the Ceriatone layout I didn't notice that the socket was rotated 180 degrees with the pin 1 triode populated first as V1a (the triode side the signal going into the amp sees first), and I wired it that way. In retrospect, I know the common wisdom is to begin with pin 6 to minimize hum and this convention is followed by many amps, not just #183 and Dumbles. I know there are references in at least two manufacturer 12AX7/ECC83 spec sheets that state hum is minimized by populating the pin 6 side first. However, does anyone have any real world experience as to how hum is affected by socket/triode orientation? I'll probably just rewire the preamp heaters, but I'm still curious. I openly wonder if there is a universal convention regarding internal wiring of 12AX7/ECC83 tubes (either in regard to nomenclature or physical configuration) that all manufacturers follow, and wonder if with some tubes it could be the case that the pin 1 side minimizes hum.

.

IMG_0238.jpg

.

IMG_0239.jpg

.

2.jpg

[bepone]

im using dc heaters for first two tubes so in this case is the same, if using ac probably is a difference and better to follow pin 7 for input.
altough humm is created with input wiring, and lead dress, from input jack, grid leak resistor and cathode circuit grounding.
to reduce humm it is better to mount first tube close to input jack.
voltgaes will drop to 470v with the tubes probably , so ok

[martin manning]

I've been moving between two circuit diagrams and two layout schematics.

Always a risky business!

I know there are references in at least two manufacturer 12AX7/ECC83 spec sheets that state hum is minimized by populating the pin 6 side first. However, does anyone have any real world experience as to how hum is affected by socket/triode orientation? I'll probably just rewire the preamp heaters, but I'm still curious. I openly wonder if there is a universal convention regarding internal wiring of 12AX7/ECC83 tubes (either in regard to nomenclature or physical configuration) that all manufacturers follow, and wonder if with some tubes it could be the case that the pin 1 side minimizes hum.

I recall seeing a note to that effect on a Mazda data sheet, but all of the other data sheets I have looked at make no such distinction, including 7025 (a low noise and hum version of 12AX7). It might be that the internal construction between manufacturers was different at some point in time, but later became standardized on a configuration that produced equal performance in the two sections. In terms of practical experience, my ODS has the 123 triode connected as V1A, and it doesn't have any hum issue.

[Richard1001]

The 678 triode is section one / the first triode of the tube.

[martin manning]

Phillips ECC83 data sheet says that with the heaters connected in parallel, the 678 triode is "more favorable" wrt hum, but is that true for all ECC83/12AX7? And is the difference significant? I ask the question because other manufacturers do not mention it.

[bepone]

find out by yourself, make full gain from the amp and change v1 tube for others . for clean doesnt matter all tubes are quiet. for max gain all is counting.

[cys]

The Mazda and Philips data sheets were the reference I was making to "at least two." They are the only ones I'm aware of. My takeaway is that I'm going to leave the preamp sockets as they are. Thanks much for the discussion everyone.

As far as the voltage goes, B+ with this PT is quoted as 420V @200mA with SS rectification. I am hoping to get about that (it was part of the PT choice).

[cys]

I was able to complete the 183 build. I am pretty blown away by the sound of this amp. It really conveys a thickness or bloom to the notes that I haven't seen with other amps. Everything is working at this point. However, the FET board needs "something" to produce more gain than it does presently. The only other thing I noticed was rather than feeding back when I hold a guitar next to the amp, the white noise baseline significantly increases when I stick the guitar a foot away from it. I don't have it in a cab yet. I purchased a head cab from Tube Town even though I'm in the States. I just got notice that customs released it so that's going to show up shortly.

As regards the FET, I'm using a 2N3823. When I was making some adjustments initially I swapped the 150K resistor for a 100k. I've got Rs as a trimmer for the time being. I'm getting 14.9V, 4.88V and 2.51V as Vdd, Vd and Vs. I want to set the FET for distortion. I'm going to have a look at the thread here, but any comments appreciated.

Also, the voltages came out to be B+ 463V, B+2 462V, B+3 440V, B+4 332V and B+5 324V.

.

IMG_0291.jpg

[martin manning]

You're on the right track for FET distortion with Vdd/Vd = ~3, but you can try different bias settings. I would keep the upper divider resistor at 150k, and adjust the lower one to get the Vdd you want, which would be around 9V I think.

[ViperDoc]

I'd love to hear a clip when you have a second. Nice build.

[cys]

I'd love to hear a clip when you have a second. Nice build.

Thanks. Once the dust settles on tweaking it, I hope to post some kind of clip.

[cys]

You're on the right track for FET distortion with Vdd/Vd = ~3, but you can try different bias settings. I would keep the upper divider resistor at 150k, and adjust the lower one to get the Vdd you want, which would be around 9V I think.

I swapped the 150K resistor back in. Irrespective of the bias settings, the FET input is giving pretty weak output relative to the non-FET one. I'm going to continue to troubleshoot until I get something decent coming out of it.

[martin manning]

Try bringing the bias back to where you are dropping half of Vdd through the drain resistor.

[cys]

Try bringing the bias back to where you are dropping half of Vdd through the drain resistor.

I've tried the gamut of biases. The one dropping half of Vdd is louder but the volume is still greatly attenuated compared to the non-FET channel. I'm starting to suspect the FET itself. I have a NTE452 on hand to swap in. Because the FET volume is so attenuated, hum is creeping in to the signal.

[martin manning]

You should be getting 18-20 dB of gain, so yes I think there may be something wrong there.