[LPBII]

I have always liked the concept of a 6V6 Plexi, so when it became apparent that the DSL15H heads are availabe used for not a lot of money I snagged one with intention of modifying it.

The word on these amps was that the green channel is too clean and the red channel is too much gain. Well the DSL 15 watt only offers those two modes that nobody used on the old JCM2000 amps. What was the thinking here?  They probably expected the customer demographic for this amp to complain if the gain was not over the top, and the clean was not clean enough. The amp can be modded, but if your not into modding amps these amps are pretty much useless to players other than modern metal players in its stock form. I will be following the mods pioneered by the guys over at the Marshall Amp forum, and the My Les Paul Forum.  I hope nobody minds if I log the journey here.
 
First impressions:
 
 Not very loud.  Even through a 4x12 and cranked up it sounded not very loud.  There is no way this amp can be used with a live drummer at this volume level.  But then I noticed that the pentode-triode switch on the back was set to 1/2 power, or only 7.5 watts.
 
 In 15 watt mode it was totally different.  It can get loud. Exploring both channels through various knob settings and with both humbucker and single coil guitars revealed the following;
 
Green Channel. Way too clean and rather boring sounding.  I had to use the Green channel gain knob on ten at all times for it to have any character at all.  Turn it down to 9 and it became very weak and thin. I normally use PAFs and single coils and do not want to back to high output pickups.

There is a tone shift button that should never be used. It makes it sound like one of those 1970s Fender Twinns that we all hated. Even with the gain on ten the Green channel sounded "plinky".  It had no sustain.  Even more troubling, was that the bass notes were boomy through the Green Back loaded 4x12. It was a like a kick drum when ever a bass string was touched, especially with the Deep Switch engaged.  The Deep Switch was not usable with the 4x12 when using the Green Channel.

 Both the tone shift and the deep switches affect both the Green and Red channels.
 
Red Channel- What a fizzy mess! Bright, harsh, and fizzy, and waaaayyy too much gain. It is basically unusable. At least for me.  Even with the gain knob at only 9:00 it is too distorted. With very low gain settings the Red is still too distorted although thinner.  With low gain settings and the volume whacked up to ten, I think I can hear some mojo from the 6V6s under all that fizz. I think.
 
With the Green channel on ten/ten it is still too clean.  And boomy.
 
What I disliked most about the amp, though, was how stiff it both sounds and feels.  It does not breathe like a traditional tube amp at all. In fact despite the all tube signal path it feels almost solid state. Stiff and tight to the point of brittleness. The Red channel does not sing, and the Green channel is plinky sounding.
 
Many have modded the Red channel into something usable but left the Green channel alone. I will be modding both.
 
 
 
 
 

[darkbluemurder]

Cool stuff - and yes, the way you describe it modding both channels seems obvious.

Good luck with the project,
Stephan

[LPBII]

Thanks,

I took the chasis out and checked the bias before running it loud at my current loud playing space yesterday.  The bias was way out of whack.  One tube was dissipating 94% of max dissipation and the other was at 82%. I set them both to 70%.  It did improve the basic tone on both channels.  The green is much less boomy now, but still needs the gain on max to sound decent. The red is smoother but still fizzy. Gone was an annoying treble spike on the Red Channel.

While I had the amp apart, I noticed that the filter caps are 100uf.  This explains the stiffness of the amp. 50uf filter caps are on the list of things to try now.

Previous to the bias check I tried it in some different cabs with Vintage 30 speakers.

V30 1x12 open back-  It sounded better than through the 4x12 for the most part. The Red channel was boxy and a bit nasal, but the Green now wasn't boomy at all with the Deep Switch not engaged. This speaker is a 90s Mesa spec that doesn't have the infamous upper mid spike.

V30 large volume 1x12 closed back- Smoother yet and a little less boxy.  Solid lows without too much boom.

I also A/Bed the amp against some others of mine for reference tones.

Green CH vs MetroAmp JTM45- Should not have done this.  The DSL Green ch sounded like a transistor radio compared to the JTM45 run clean. 

Red Ch vs Silver Jubilee 50/25 watt. I expected this. The Jub completely destroyed the DSL's gain channel. The Jubilee gain channel just sings and is so 3d and dynamic. The Jubilee has not near the hiss and buzz when its not playing a note either. The DSL red Ch is real noisy.

Orange Micro Terror 20 watt.  Should not have tried this either.  The 1/2 solid state Orange sounded way more tubey and organic. And dynamic. Although the Orange only has three knobs you can set the gain knob up to about 3 O'clock and control the tone from your guitar volume. Turn your guitar down and the Orange cleans up (not completely of course) and crank the guitar volume on a humbucker and its a nice driving overdrive with out fizz. It is as loud as the DSL15.

At the rehersal space, which has high ceilings and hard wood floors and plenty of natural hall reverb, the DSL sounded pretty good. The fizz gets lost, for the most part, in the mix and also the woosh of the room.  You can still hear it but with the masters cranked up, you can hear the 6V6s starting to take over and overide the fizz and thinness.  Below 6 on the masters and it is noticably thinner. The triode mode cranked is good on both channels. I ran it through the V30  closed cab.

 

[darkbluemurder]

While I had the amp apart, I noticed that the filter caps are 100uf.  This explains the stiffness of the amp. 50uf filter caps are on the list of things to try now.

Are you saying 100uf also for the preamps? Way too much IMHO. For the preamp 20-22 uf is sufficient. I don't mind 100uf on the power tube plates and screens but 50uf mostly works.

Green CH vs MetroAmp JTM45- Should not have done this.  The DSL Green ch sounded like a transistor radio compared to the JTM45 run clean. 

The JTM45 is one of my favorite clean tones - hard to beat.

R

ed Ch vs Silver Jubilee 50/25 watt. I expected this. The Jub completely destroyed the DSL's gain channel. The Jubilee gain channel just sings and is so 3d and dynamic. The Jubilee has not near the hiss and buzz when its not playing a note either. The DSL red Ch is real noisy.

It is long ago that I had a Jubilee (the 50W top). I should have kept it. It sounded really nice and was not as loud as the regular 50W Marshalls are.

Cheers Stephan

[LPBII]

All the caps are the board. These caps are on the power amp, and the pre-amp derives its power supply from there through a resistor instead of a choke.

50 watt Marshalls have always used 50uf filter caps through out, including 50uf on both sides of the choke.  The original JTM45 only used 16uf . Many run a 32uf on the power amp side of JTM45s (including mine) to prevent ghosting.

The 100 watt circuits have general ran 100uf on the power amp side, and 50uf on both sides of the choke like the 50 watt, but there is quite a complicated history of the evolution of the filtering of the 100 watt amps. Early plexi 100s had low filtering (often ghosting) and then by about 69 they had high filtering.  Then in about 1986 they reduced the number of the filter caps on the 100s to make them easier to build.  A lot of people think that the pre 86 100 watt JCM800s have more punch.

Hiwatt 100s make an interesting comparision. Hiwatts ran 200uf on the power amp, 100uf on the screens, and 50uf on both sides of the choke. (BTW the Jubilee tone stack and taking the signal to the tone stack from the plate instead of the cathode are based on the Hiwatt amps design)

 Orange and Matamp ran 100uf  and only 16 uf on both sides of the choke.

Fender Tweeds ran about 20uf caps throughout. The Black Face Fender amps started running about 70uf on the power amp side which became more standard after the CBS take over going into the silver face era. By the 70s many Fenders were running 200uf on the power amp side.

[LPBII]

I can't seem to find any schematics for the DSL15 online.  Plenty of JCM2000  DSL schems but the boards and lay outs are different. That makes it it kind of a puzzle to find the components affecting the green channel.  The Red channel mods are well known.

The plate voltages are up around 425V,  which means I can't use old stock 6v6 tubes pulled from an old radio or something.  The max rating on old style 6V6s are 315v and 12 watts max dissipation.  Apparently new design 6V6s can handle up to 450 volts and have 14 watts max dissipation (and sound more 6L6 like).

[LPBII]

Fixing the Red CH might be as simple as using a 12DW7 pre-amp tube in V1. The 12DW7 has a 12AX7 triode for the A triode ( pre-amp tubes are actually two tubes or triodes in one bottle) and a 12AU7 for triode B. A 12AU7 has only 1/5 the gain of the standard 12AX7. So if the Red Ch uses V1B, that will be a far easier fix that modding the printed circuit board.  Still need to up the gain of the Green Ch in that case, though, because the Green Ch sux if its master volume and gain are not dimed. If the Red Ch uses V1A then it won't affect the Red Ch and will totally gut the already gutless Green CH. 

[darkbluemurder]

It certainly does not hurt to try lower output preamp tubes, especially the 12DW7 came to my mind as well. The 5751 would also be a good candidate.

Cheers Stephan

[LPBII]

The main problem with using lower gain pre-amp tubes, is that I don't want to lower gain on the Green CH at all. In fact I would like to increase it so it behaves more like the Green Crunch mode on the big DSLs.

Just to clarify the signal chain of the DSL:  The guitar signal goes to one side of V1. After that the Green CH signal goes on through both V2 triodes and into V3A and then into V3B. The signal is taken from the cathode of V3B to the tone stack (EQ). The amplitude of the Green signal from V1 to V2 will determine how clean and how much gain the Green Ch will be/have. 

The Red CH signal is split from the Green CH signal right after V1 and re routed so that that it cascades into the other V1 triode at a clipped level. It then cascades through both sides of V2 and V3, making it five cascaded gain stages at overdriven levels.  But I really need to find a Schematic of the 15 watt version.  The JCM2000 schem of the DSL preamp design is almost unreadable, it so convoluted and complex. The DSL15 only has Green and Red modes, without crunch, and lead 1 modes.

I have seen some internet postings that 12AU7 side of a 12DW7 actually pins out to the Green CH when used for V1. That would be exactly opposite of what I want. JJ makes a 12DW7 with the triodes reversed, but I don't know how easy it is to obtain that tube in the USA.  It might be just as well to mod resistors and capacitors on the board......

The nuclear option is to just gut the amp and use the chassis and transformers to support a 6V6 Vintage /Modern type build.

[LPBII]

The project is kind of at a stand still until I can order parts online.  Local mom and pop electronics supply stores don't exist anymore.   I don't want to place an order until I find out what I need to up the gain on the green channel. I may have to trace the traces on the PCB to draw a schematic, because I have had no luck in locating a schematic.

In the meantime, I read that a 12AU7 lowers the gain on the red channel down to a crunch. That would be not enough gain on the red channel for soloing, so a 12DW7 (reversed?) may not be the ticket anyway.

Playing the green ch with a ToneZone and Bluesbucker loaded Charvel sounded good (gain still dimed).  Obviously they have engineered the green Channel for high output pickups.  This of course makes sense because that's what most people play with these days. I don't though. I play mostly vintage type single coils and PAFs these days.

The green channel sounds better through a open backed V30 loaded cab than through a closed back cab.  It breathes some life into the clean sound if the room is a little dead. This amp could have used a reverb at least for the green ch. 

[LPBII]

Still have not made in mods (been too busy) but its apparent that DSLs are wicked bright.  Just to get the JTM-45 into the same range required diming the treble and presence knobs. And the Red Ch is even brighter than the Green Ch.

[LPBII]

I finally found a schematic of the DSL-15H:
 
*The signal is split between V1a and V1b with the red channel driving directly into V1b to create distortion. This signal path dead ends before V2 unless i's switched in.
 
* The Green (clean ) Ch signal path by- passes V1B but does not go straight to V2.  There is a Fender Black Face style EQ section or tone stack (it's fixed internally by the resistor and capacitor values with no adjustment knobs). Moreover, this tone stack is located, as in a Black Face Fender, right after the very first gain stage.
 
 *From there, it's on to V2. Once again the Green Ch signal path is dead ended unless it's switched in. Either the Red or Green Channels are switched in just before V2- at the expense of the other.
 
*V2 represents the extra tube we would find added to a modded Marshall JCM800 2203, or the typical modern high gain tube amp.  Apparently V2A will overdrive if the signal it receives is strong enough (as in the Red channel signal). V2A cascades into V2B.
 
And V2B also cascades into V3A.
 
V3B is a cathode follower that goes to the master tone stack, or EQ controls.  This is the amp's master EQ section for both channels-although the green channel is pre-EQed
 
After the tone stack are the dual volume controls (for the Red or Green Chs switched in and out with them) and it's off to the PI and output tubes.

OK, how do I mod the Green CH more to my liking? On the big DSL's the crunch mode essentially by-passes the pre-EQ.  The DSL15 doesn't have the alternative circuitry that the signal path for the crunch mode would follow. I could add these on a separate bread board, but all I really want is a little more gain.  It looks like I can just change the value of R 76 to get that. I may need to change the slope resistor for the green ch pre-EQ tone stack as I suspect that is why it is so bright. 
 

[darkbluemurder]

What is the function of R76 - is that the fixed resistor for the midrange? You could increase that for a boost effect. To tame the brightness you could add a capacitor in parallel to the resistor. The older Mesa Boogie Mark IIs have a 33k in parallel with a 0.005uf cap in between the ground connection of the mid control and ground. This can be bypassed (for normal operation) or inserted as a "gain boost". For less gain boost, decrease the resistor.

The Hughes & Kettner Puretone has a 1 meg pot between the ground connection of the mid control and ground, labelled "Growl" - turned up it is like lifting the tone stack as in the Mesa Boogie Mark I. This is quite a radical boost.

Cheers Stephan

[LPBII]

Thanks for the tips and all your comments Dark Blue Murder.

I think the mid range setting resistor in this case would be R75

[LPBII]

I found a well stocked electronics store and have acquired everything I need-I think.  I haven't had time to remove the board yet, though, and I'm busy replacing filter caps in another amp.

I did pull a 2x12 with G12T75 speakers out of storage.  Not good. It accentuated just about everything I don't like about the stock amp. I need to play through G12T75s every once in awhile to remind myself why I don't like them.