Marwatt build thread…

[Raoul Duke]

Ahh, that’s what I’m missing. Thanks for the reminder Martin. When I stop looking at this for a few months I forget these important details.

I need to get the current down to the 15-30ma range - so my 4k7 in the bias circuit needs to go up.

Or… might I be better off increasing the first dropping resistor to 1k from 750R? Let me take a few more measurements…

[Phil_S]

If the schematic is accurate, there is a voltage divider using a 2.2m resistor. Adjust the voltage divider to bleed off move voltage.

[Raoul Duke]

Thanks Phil, I appreciate the help!

So I took more measurements and don’t think the main dropper needs to change - so I swapped the 4k7 bias resistor for a 2k7 and she biases up right around the center of the pot travel.

Tried to get sound and began with the Marshall high input. Volume works, tone works (although not a lot of variation - might look into voicing later). Master volume works fantastic as well; doesn’t seem to change the tone noticeably wherever it’s set.

Moved on to the Marshall low - nothing.
Then went to the Hiwatt low - nothing.
Last - the Hiwatt high - nothing.

So I guess it’s time to either get the signal tracer or the scope on it and see where I went wrong. Never used these Rean jacks before, so maybe I’ll start by looking at what I did there. I’ve worked on this piecemeal over 3-4 months, so my usual focus has been hard to maintain.

I also see that B+4 & 5 don’t vary at all and the PI voltage strikes me as odd.

Here’s the numbers:

AC in: 120vac
Heaters: 6.7vac
B+: 373
B+1: 367
B+2: 352
B+3: 352
B+4: 330
B+5: 330

V1 plates: 170
cathodes: 1.2

V2 plate A: 197
cath. A: 2.0
plate B: 200
cath. B: 1.8

V3 plate A: 208
grid A: 116
plate B: 204
grid B: 116
cathode: 140

V4 & 5 plates: 374 idling at 20 mv at the sensing resistors - which seems good anyway.

Back to the drawing board and schematic.

[sluckey]

I also see that B+4 & 5 don’t vary at all and the PI voltage strikes me as odd.

Here’s the numbers:

AC in: 120vac
Heaters: 6.7vac
B+: 373
B+1: 367
B+2: 352
B+3: 352
B+4: 330
B+5: 330

Dude, there is no B+5! I think what you are calling B+5 is actually B+4. There is only a piece of red wire between the two points so it's all just B+4. That's why they will always be the same.

[Raoul Duke]

Thanks Steve; not sure why I didn’t see that.
Since the reference is “battery plus” - I understand that it’s successive nodes in the power supply (to your point) - requiring a filter cap and dropping resistor for separation. For some reason I was focusing on plate load resistors.

All that said - the B+4 voltage seems high based on the schematic, but the tube voltages seem within a normal range to me. Is the B+4 not the target; but rather the tube voltages? I suppose that could be part of the final “by ear” adjustment once everything else is working?

I always appreciate the course corrections - It’ll all sink in eventually, lol.

Thanks again!

[Raoul Duke]

Reviewed the schematic, traced everything in the pre-amp - everything looks good to me; so I hooked it up to the scope.

Consistent with my play test - I see signal coming through to the grid of the Marshall Hi channel; but none getting through anywhere else.

This has to be the jacks then I would think? I’ll have to really pick apart my wiring there and see what’s going on. I grounded the shielding to the jack ground tabs which all connect and run to the same pre-amp ground lug - so if one channel is grounded - they all must be, right?

Oh, wait a minute - I think I figured it out…

[Raoul Duke]

Yep, just as I suspected. I wired the inputs upside down (confused the hot side with the switched side). First time using Cliff-type jacks in a build - but that’s how I learn, lol.

When I first started getting a clue - I googled for a reference just to be sure I was barking up the right tree. Wouldn’t you know it - sluckey strikes again! Found this excellent graphic reference created by the man himself and duly stuck to the wall next to my bench:

IMG_0303.gif

The amp sounds pretty good. The Hiwatt side is definitely interesting and like nothing else I have. Really responds to the mid knob as a tone and gain shaping control. The Marshall side is kind of “meh”, but given all the info out there in this circuit - I’m sure I can tweak it to my liking. Haven’t tried jumping the channels yet. Glad I didn’t use the EL84s as well. Now for the tube rolling. Going to start with using a 12AX7 in V2 (over the 5751).

Runs nice and quiet, too. Wasn’t sure what to expect with used Weber transformers - but nothing negative to report so far.

I really appreciate all the assistance and coaching to get to this point. Always learning more with each project and couldn’t do it without the help here on this board. Very much appreciated!

More to follow…

[Raoul Duke]

The Marshall side wasn’t really impressing me much; and truthfully - part of my motivation for building this circuit was to get a better understanding of how to voice a pre-amp; so… After a couple of days listening and researching, here’s the changes I’ve made:

V1:
- Plate load resistor from 100k to 150k
- “A” triode bypass cap to 4.7uf, parallel resistor remains 1k5
- “B” triode bypass cap to 10uf, parallel resistor to 820r

This gave me a little more mid crunch and brings the volume of this channel closer to the Hiwatt side. Real happy with the way it’s sounding.

Last little weird thing is that on the Marshall channel, when bringing the tone pot past 3/4 of the way up - the channel seems to lose a little volume. My research suggests that this is a perceived change that is actually the mids and bass being reduced to such a point that it sounds like volume loss.

My question is: is this in fact what’s happening and is there a way to mitigate that - or is it an inherent characteristic of this circuit?

Can’t seem to find a conclusive answer so I thought I’d ask here.

Thanks!

[cdemike]

Really glad this build came together!

Last little weird thing is that on the Marshall channel, when bringing the tone pot past 3/4 of the way up - the channel seems to lose a little volume. My research suggests that this is a perceived change that is actually the mids and bass being reduced to such a point that it sounds like volume loss.

My question is: is this in fact what’s happening and is there a way to mitigate that - or is it an inherent characteristic of this circuit?

Can’t seem to find a conclusive answer so I thought I’d ask here.

Have you measured the AC potential at the output jacks/speaker when this happens? I suspect one of three things is happening:
- Turning the volume up to 3/4 or higher induces oscillation.
- The speaker(s) you're playing through are starting compress at those volumes making the amp read quieter than it is by virtue of it reading as less dynamic;
- The power section may be getting so compressed at those high levels that the amp reads as quieter than it is (same effect as with the speaker);

The AC potential test would help rule in/out the first scenario. If there's no oscillation, playing into a different cabinet/speaker would narrow it down to the third scenario. If you get there, I'd suggest biasing the amp cooler (50% dissipation works well for me when I want wider dynamic range).

[Raoul Duke]

Thanks Mike!
I’ll certainly try that and post results (to include the bias change).

Trying different speakers out today and so far my WGS ET-90 seems to sound the best with this build. Thinking about pulling the trigger and making a Hiwatt style cab with Fanes, Reeves, or Weber Thames - but trying to find one that sounds good already here.

Always appreciate the input!

[Raoul Duke]

After researching all the idiosyncrasies of the Marshall 18w normal channel and trying to solve the weird “dip” at 3/4 in the tone control as well as re-voice it into something that sounds good - I scrapped it.

My “Marwatt” is now a “Dewatt”. I rebuilt the channel with a 5E3 tone stack and it is sounding much better to me now. The dip is gone and the channels are more evenly matched volume-wise. I’ll fix my layout and post it for anyone interested.

Thanks again for all the help guys!