D'Mars ODS build completed

[MakerDP]

So I finally built a D'Mars ODS. Been wanting to do one since I first found this site.  I used the "5879 & FX" schematic found in post #2 here...
http://el34world.com/Forum/index.php?topic=17305.0

I really really like the sound of this amp. It really matches with my playing style very well, which is an electric blues/funk sound (greatly influenced by Robben Ford, Freddie King, and others like that) and my PRS SingleCut with their #7 PAF-style pickups. I initially tested it with a Vintage 30 in a closed-back cabinet and it was pretty bright and shrill. Then I tried my WGS ET65 (their version of a Celestion Heritage 65) in an open-back cab and that speaker is a MUCH better match for the amp. Tames the highs a bit and has a nice big, open bottom. The open-back pine cabinet has a really nice "3D big" sound that really fills the room nicely.


I deviated from the design in three ways:
1) I made my own layout so I could use my preferred grounding scheme. The only reason I did this is because I am used to this style so it makes building and debugging easier for me. Plus, making my own layouts really helps me wrap my brain around how the circuit works.
2) I ended up bypassing the effects loop altogether because I didn't like how it dramatically colored the tone and distortion level in the amp, especially in Clean mode. I actually left the "FX Level" control there but added a 100k resistor in series between the control and the PI input, which in reality is really nothing but a pre-PI master volume control.
3) I used solid-state rectification. I started out going with the valve rectifier but I think my rectifier tube was bad from the get-go. I checked all my wiring and was correct but since I blew the rectifier pretty much straight-away on a Friday evening, I didn't want to get derailed waiting for a new one so I just wired in some diodes for a solid state full-wave
rectifier.

The amp fired-up nicely first try (except for the rectifier incident of course). What a good feeling - there was a LOT of time spent on this amp! This was also my first "channel switching" build and with three relays I was worried about all that, but it went smoothly once I figured out I had to reverse the polarity on my LEDs in the footswitch. There is a helpful thread about that here somewhere. I think there is a link to it on the D'Mars schematic page.

Only one issue I need to track down... when the PPIMV is turned up past seven or so, I get a really loud low-frequency oscillation/feedback that I need to track down but the volume level at that point is already borderline too loud for most situations I will ever be in - still plenty loud enough for small to medium-sized venues but I need to track it down and fix it anyways.

My hat is off to Tubenit for this design. It really does deliver! Many thanks to him for both freely sharing his design and for his help in the correspondence I have had with him over the last year or so leading up to this build!

Hoping to have a full build-report with pics and sound samples on my blog soon.

[sluckey]

when the PPIMV is turned up past seven or so, I get a really loud low-frequency oscillation/feedback

Sounds like NFB phase issue. Reverse the OT primary plate leads to correct. If oscillation gets worse swap the plate leads back.

[tubenit]

I think Sluckey's suggestion might be the answer for you.

I really really like the sound of this amp. It really matches with my playing style very well

Glad you like the amp tone!   Will look forward to your soundclips and photos.

I am curious about the FX changing the clean tone for you?  Did you use the 470k/.001 to make it a parallel FX loop?  And are you using a 12AY7 in the loop?  I typically use a 12AY7 or 12AV7 there.

With respect, Tubenit

[MakerDP]

when the PPIMV is turned up past seven or so, I get a really loud low-frequency oscillation/feedback

Sounds like NFB phase issue. Reverse the OT primary plate leads to correct. If oscillation gets worse swap the plate leads back.

Thanks Sluckey. That was going to be the first thing I tried. I'll swap them here in the next hour or so and report back.

[MakerDP]

I am curious about the FX changing the clean tone for you?  Did you use the 470k/.001 to make it a parallel FX loop?  And are you using a 12AY7 in the loop?  I typically use a 12AY7 or 12AV7 there.

With respect, Tubenit

Yes I included the parallel bit and used a 12AY7.

The problem I was having with it was that the "FX Rtn" pot (not the FX Lvl pot - like I said, that's really just a Master Volume control) was always active, even with nothing plugged into the loop, basically turning the FX recovery triode into another gain stage in the main signal path. It was nearly impossible to get a clean tone that had any volume to it at all. When it is set to "0" it also mutes the whole amp. (Maybe I have a wiring issue there, but I went over it multiple times.)

Looking at the design of the loop, I am inclined to think two things:

1) The FX Send and Rtn jacks being tied together via their switches. Looking at the schematic, doesn't that force the signal to always go through the loop? Most loop designs I have encountered have no switch connected at the "send" and the "return" switch is connected to ground so the loop's return stage is silenced when nothing is plugged in. But, that also leads to #2...

2) I think I can keep the loop and my issues with it will be solved if I move the insertion point of the "parallel path" to the other side of the FX return triode - actually to the "FX Lvl" (MV) pin 1/Lead Relay junction. Combining the jack switching from #1 and this the triode will only be in the "FX" portion of the loop and completely bypassed and shunted to ground if nothing is plugged in and the send and return pots will have no effect on main signal path at all.

So, first I am going to swap the OT leads to try and fix the oscillation since that is a pretty easy fix. Then I will try and do the jack switching layout as in #1 and move the insertion point of the parallel path of the loop as in #2.

I'd appreciate any insights you or anyone else has on this FX loop.

[MakerDP]

when the PPIMV is turned up past seven or so, I get a really loud low-frequency oscillation/feedback

Sounds like NFB phase issue. Reverse the OT primary plate leads to correct. If oscillation gets worse swap the plate leads back.

Thanks Sluckey. That was going to be the first thing I tried. I'll swap them here in the next hour or so and report back.

Yup. That fixed it! 

[tubenit]

I am curious about the FX problem?   Double/triple check to make sure it's wired right.  I never had a problem getting a clean tone with the FX.  In fact, one of the things I like about the D'Mars is that it had more contrast between the clean and the overdrive then other "D-inspired" amps I've built.

Check out the clean tone on the D'Mars at the start of this sound clip compared towards the end of the soundclip.
http://www.soundclick.com/player/single_player.cfm?songid=11842172&q=hi&newref=1

I think on my Tweed BluezMeister with an FX that I might have a 82k resistor to ground instead of a return pot and I also think maybe the end of the parallel loop  (470k/.001) re-inserts into the grid of the next triode?  I don't have the schematic in front of me right now?   

I always have something in the FX loop.  Either delay or reverb all the time.

With respect, Tubenit

[sluckey]

Now that the NFB issue is sorted out you may want to re-evaluate the FX loop. That NFB issue may have tainted your opinion of the FX loop.

[MakerDP]

I've checked it in the layout and in the wiring multiple times.

The triode is always in the signal path. Therefore, it alters the gain of both the processed signal and the un-processed signal.

This is probably just a difference in FX loop design philosophy. Some prefer them to be as transparent as possible, others don't mind if they color the tone. Each viewpoint is only correct to the person with that opinion in this case.

Maybe i'll wire it with a temporary switch so I can try it with and without the loop for an easy A/B comparison on-the-fly.

[MakerDP]

Here are my filter cap voltages, just as a sanity check. They all seem pretty reasonable...

A 373V
B 368V
C 340V
D 243V
E 235V

Nothing there jumps out at me as being problematic.

[tubenit]

I am a little confused in that I always have thought of the D'Mars FX as being pretty transparent?

Please consider trying this.  Remove that triode  50uf/1.5k and replace it with a unbypassed 2.2k.   How does that sound to you?

Then try the FX send and the FX return around 4 and the FX level around 6-7.  How does that sound to you?

With respect, Tubenit

[MakerDP]

Sorry I let this drop. I had a big show last weekend and I wanted to use this amp. Since it was working well with the loop bypassed out of the circuit I didn't want to mess with it and possibly mess something up for the weekend.

Everything sounded great to my ears at stage-level volumes with my reverb and delay in front of the amp and I received quite a few compliments on the sound. I enjoyed having a "pure tube-overdrive" tone where I did not have to rely on any pedals for distortion at all. The clean channel is dead-quiet and the OD channel only has a very low-level hiss that you would normally hear in a higher-gain situation, but even that it is still pretty darn quiet. I was VERY pleased with the noise levels in this amp because it is my own custom layout and is by far my most complicated build to-date. All the relay switching works great too. I really like the mid-shift on the clean channel. It can really "fatten" up the tone when that's what I want. I pretty much never use it on the OD channel so far though.

I also put a Pearly Gates pickup set in my PRS Singlecut before the show. Man that bridge pickup sounds awesome with this amp on both channels. It was a great upgrade from the stock #7 PUPs. The neck might be just a tad too hot-n-gritty for my tastes though so i might go with a '59 or a Seth Lover or back to the stock #7 in that spot.

I may try your suggestion. Going non-bypassed on the cathode resistor would certainly reduce the gain of the recovery stage. I'm just not sure I even want to use the effects loop since it sounds so good without it  and I am thinking "less is more" might be a good way to go here. Taking out the loop and going with an IRF820 for the second CF would allow me to eliminate a whole tube altogether. Or I could go MOSFET CF and use that extra tube for a one-tube reverb circuit instead. I did notice that around 6 to 8 is as far as you can go on what you call your "FX Level" control before it allows enough signal through to start introducing a bit of power-amp distortion but it plays well with the PPIMV so it can still be loud and clean at the same time. (You really should start calling that a Master Volume.)

But all-in-all a great design! Loving it!

If anyone else is on the fence about trying it they should go for it. It's not for the beginner though... complicated build for sure.

[tubenit]

Very well stated review!  Thanks for the detailed descriptions. 

I think it's worth trying the non bypassed FX return prior to giving up or changing that design.  All you have to do is remove the cathode cap & if you don't like it, go another route. The delay in the FX loop will be MUCH quieter then into the front end of the preamp.

Nice to hear you got compliments on the tone!   I found the D'Mars to have the best "difference" between clean and OD of any amp I've built and it makes for a very versatile amp, IMO.

My D'Mars was very quiet also. I would be interested in knowing more about what you did for layout and grounding?  Any photos or layout drawings would be appreciated. 

With respect, Tubenit

[Willabe]

.....but it plays well with the PPIMV so it can still be loud and clean at the same time. (You really should start calling that a Master Volume.)

It is a MV, calling it's a PPIMV (post phase inverter master volume) is just saying it's post not pre PI.

They can/do sound different in different amps.

[MakerDP]

.....but it plays well with the PPIMV so it can still be loud and clean at the same time. (You really should start calling that a Master Volume.)

It is a MV, calling it's a PPIMV (post phase inverter master volume) is just saying it's post not pre PI.

They can/do sound different in different amps.

No I'm talking about what he is calling "FX Level" is in reality just a simple pre phase inverter Master Volume control. I've built amps with both types before. They do most definately have their own affect on the sound.

[Willabe]

No I'm talking about what he is calling "FX Level" is in reality just a simple pre phase inverter Master Volume control.

Sorry I missed that.

[MakerDP]

No worries. I can see how you would think that the way i worded it.

[Tone Junkie]

I don't hardly use the PPIMV anymore. I like a tube driven effects loop in all my amps. so I just use the effects level as my MV.
Bill

[MakerDP]

I am really starting to just absolutely love this amp with my PRS/Pearly Gates guitar. I've been spending some time dialing-in my sound with it and I am getting really comfortable having it behind me on stage. I've used it for two shows now and I have not experienced that "ugh something's totally wrong" feeling that you usually get when you are using something different that you are not used to on stage. On the contrary, I love what I hear. I've never really been a fan of "stomp box distortion" and now I am actually considering taking my Plimsoul off my pedalboard because the overdrive from the amp is SOOOO much better than what I can get out of that pedal (which is no slouch by the way) and I have not used it at all since building this amp. I am still using my EH Soul Food as a transparent clean boost for pushing clean-channel leads into that bluesy-overdrive territory, which the two create very nicely together.

I think I am going to just leave the effects loop bypassed. I am totally happy with my effects reverb and delay out front. I don't use a ton of delay so it's no biggie to me. Maybe someday in the future when I have some time and I'm bored I will revisit this but for now I am totally happy without it.

I am curious Jeff... how much have you experimented with the resistor that sets the screen voltage on the 5879? Have you tried lowering the voltage on the screen at all? I am wondering if I might be able to coax just a little more touch-sensitivity out of the OD channel if I lower the screen's voltage just a wee bit. But if you've already tried that and didn't find much difference I won't bother messing with it.

I have not experimented with rolling other tube brands through the amp yet either. Right now I am using Tung Sol reissue 5881's, a Phillips for the 5879 and JJ's everywhere else.

I am working on some recordings of the amp too. I was hoping I could have some very high-quality video of my band live at our last show but the sound tech screwed-up my guitar signal going to the video feed so it was not even in the mix. Such a shame because the quality of the rest of the mix and the video itself were outstanding. Good thing we weren't paying for the video!

Based on messing with some settings on other guitars, I am thinking a Tele would sound wonderful through this amp too. Man I really need to just bite the bullet and buy one already -- then beg for the forgiveness of my wife.

[tubenit]

how much have you experimented with the resistor that sets the screen voltage on the 5879?

I haven't experimented with it. I am curious how much of a difference it would make?

I am glad you are enjoying the tone of the amp!  Very cool.

With respect, Tubenit

[MakerDP]

Thanks. If I get some time this week I will start experimenting with it a little bit and report back my findings. I've been pouring through the chapter about pentode preamp tubes in Merlin's preamp design book and he mentions that with pentodes, it's the screen voltage that has the biggest affect on distortion and tone. He touches on it a little bit in his website's pentode article but his book goes much more in-depth.

Kind of flying blind with the 5879 though because any datasheet I can find for it is a little incomplete in the graphs department so it makes it a little hard to draw any relevant load lines for any chosen screen voltage. Pretty much just test a few slightly higher value screen dropping resistors and see what happens.

[MakerDP]

I thought I would update this thread on where I stand with this build...

I have not had a chance to mess with the effects loop yet. It remains bypassed for the time being. I will get to it though.

I decided to try un-bypassing the clean channel tonestack when in OD mode. It sounds really nice to my ears. I like being able to adjust the tone a little more in OD mode. I have not really noticed much difference in the quality of the OD sound or feel so I think I will leave the full tonestack in the signal path for now and switch it back after a bit to see which I like better.

I just got a mint-condition G&L ASAT Classic Tribute Series in Candy Apple Red off Reverb.com for a steal.... $425 shipped with a hard-shell case... what a gorgeous looking and playing guitar! This guitar really shines with this amp. It's like they were made for each other. This guitar kind of sounds like what would happen if a tele pickup and a P90 had babies... a tele with a lot of "growl" which mates very nicely with the OD channel of the D'MARS, especially in the middle and bridge position.

I am also thinking of going switchable fixed/cathode-bias on the amp just to see how it will perform with fixed-bias while making it easily reversible if I like it better in cathode-bias mode and to have the flexibility of either mode.