S.I.R. 100W SuperLead Schematic pt. II

Everything from original vintage Marshalls to reissues.

Moderator: VelvetGeorge

Post Reply
User avatar
CoffeeTones
Senior Member
Posts: 1112
Joined: Mon Oct 11, 2010 9:52 pm
Just the numbers in order: 7
Location: USA

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by CoffeeTones » Sat Oct 27, 2018 4:55 pm

I'd like to see your new scheme to be able to discuss it with you more accurately. If it were my amp, I would remove the .047u from the input and adjust the grid stopper 22k or larger, I'd try a 3n3 bright cap, a 35k - 39k slope to manage highs and bass response, use 470k/470k//500p divider. Adjust the pre PI node dropper to get the node at or between 330V to 370V and make your choice there as mods progess. 100k, 150k, 220k or 270k stage 1 plate resistor to taste(once PI voltage dropper is set), 10k to 22k PI tail(which will affect PI voltages), Choose between 820r or 1k cathode on stage 4 (once PI voltage dropper is set) - that will affect preamp voltages. PI and all stage voltages tie to each other. Once you learn how important voltages are with specific changes, it will be easier to relate them and adjust as you go. Adjust guitar pickups down or up 1/4 turn at a time to clear things up and manage attack. If they are slightly too close to the strings, the sound will be trashy.

User avatar
CoffeeTones
Senior Member
Posts: 1112
Joined: Mon Oct 11, 2010 9:52 pm
Just the numbers in order: 7
Location: USA

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by CoffeeTones » Sun Oct 28, 2018 10:27 am

I looked over your schematic. IMO, the .047u at the input should be removed or replaced with a .1u but connected in a different way. I'd connect the .1u to the input jack tip, connect the 1M load from the output of that .1u to ground, at that 1M / .1uf junction connect a 22k or higher stopper to the V1 grid. You can still use the hot shield and the input will be protected from B+.

You have a lot of smoothing with snubber caps, hotshields and grid resistors in several locations but you are wanting more highs. I'd start by reducing the 1n plate bypass and / or the 180k stopper. You are already getting a grid stopper effect from the 1M pot so I'd remove or reduce the 5k6 stopper to 820R. I might also choose to move some of the plate bypass / snubbing to the V3 plate or across the 100k, V3 load and / or remove the second hot shield which has proven to be too much by most who have tried multiple hot shields.

I'd use 22nf as CC off of V1A. The 2n2 may be fine but might not be great with some changes. I'd try 3n3 as bright cap. 33k - 36k slope. 35k to 36k will tighten bass but not make the highs so prominent.

The switch at the 470k divider is doing nothing as drawn in your scheme, but I'm sure that is just a mistake in the schematic. I'd replace those two 470k resistors with 560k and adjust the switched one to taste.

I'd connect the NFB to the 8R tap to slightly tame bass and make the highs more clear and articulate, but not take the life and bounce out of the amp. That will also help compensate for the loss of NFB when the PPIMV is turned down. I'd install 2m resistors on the PPIMV because 220k output grids give the best sound, output and response with 22n or 47n PI output caps, with EL34 type tubes. 150K, I'd save for the 6550 tubes, even then I might increase the PI output caps with 6550s to get the time constant closer to the 22n - 47n / 220k combo.

I'd adjust the pre PI dropper in the power supply to 12k - 15k in order to bring the PI node voltage up, and / or reduce the 10k dropper prior to the CF node, which will increase the voltages to V1, V2 and V3. I'd shoot for 185v - 195V at V3 pins 1 / 7 and 1.1V - 1.3V at V3 pin 3. Pin 3 at V1A should be 1.8V to 2V.(you may have to adjust another dropper to get to that voltage) The difference between the node and plate of V2B should be close to 30V for the best breakup and response I've found. You can always play around with the PI node voltage between 330V and 365V after other changes, to find your preference but watch what it does to preamp voltages and preamp tube bias. PI plates sound best at 220V to 235V. A stock JCM sounds great around 330V at the PI node and many of these modded amps sound great around 345V - 365V at that node if the preamp dropping is dialed in and cathodes biased correctly. These voltages can also dictate whether of not you have start putting bandaids in the circuit to control bass, highs and feel.

Your filtering is good but I'd use a 3H choke. IMO, your voltages are close but V3 is low and the PI node is a little low.

You have a lot of peaking at V2A and V3A cathodes, then a lot of snubbing in multiple places. I'd think about backing down on the peaking and snubbing to make the amp sound more natural. The snubbing / smoothing you have now really comes through on the recording and does not sound natural to me. IMO, the closer you stay to a stock JCM 800 with an added stage, the better your amp will sound. Voltages in stock JCM 800s can vary depending on which PT they were equipped with and that affects the entire amp. As always, just my suggestions from experience. Take from it what you wish.
:rockon: :toast:

BRMSlash
Senior Member
Posts: 479
Joined: Sun Jul 18, 2010 9:00 pm
Just the numbers in order: 13492
Location: Australia

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by BRMSlash » Mon Oct 29, 2018 12:45 am

Thanks CoffeeTones,

I did the 47n input cap that way as the 1M resistor was already soldered across the terminals of the jack.

When I was using the NexGens, they were so bright I had to snub & smooth a lot. I'll have to try re-work the snubbers etc to suit the TAD's.

2n2 1st stage coupling cap was used as I was trying to cut bass with the 1u electro cathode bypass. 2n2 was as a low as I'd go to keep the bass tone decent. The TAD mustards seem to have a more gradual bass roll off, so seem to work were other caps may not.

I can only get 33k or 39k carbon film resistors that I like the sound of. 33k seem to sound more correct than 39k to me in the tone stack.

The switch is there only for tweaking. I'd set it up to switch bright cap & resistor to ground to maintain a more consistent gain with different bright caps. As it is, yes, the switch is not really needed as the resistor to ground is 470k for both bright caps. Please note that I haven't tweaked for the 100p bright cap for a while, so 470k may no longer be correct.

Changing the NFB tap is a good idea. I haven't thought of that for quite some time.

Yes, generally I try to aim for the highest voltage in the B+ line within ECC83 specs. It looks like I've lowered the voltage at some stage, so should be able to bring it back up a bit. I'll have to review.

Thanks, I'll try removing/lowering some snubbers & peakers to suit.

Cheers

User avatar
CoffeeTones
Senior Member
Posts: 1112
Joined: Mon Oct 11, 2010 9:52 pm
Just the numbers in order: 7
Location: USA

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by CoffeeTones » Mon Oct 29, 2018 2:53 am

You can try 100k // 56k for the slope to reach about 36k

The Marshall 2555 schematic shows the .047uF connected at the input in the way I described the .1uF. I have not connected it as you have but as I understand it, the .047uF applied in that way affects the load and tone. I've read of people changing the value of the .047uF cap to alter the effect. I've also read of it causing noise in some instances. Using .1uF the way I described should not affect the tone or load in that way, but protect the input from B+. Notice how the 1M load comes after the cap in the 2550 and 2555.

http://www.drtube.com/schematics/marshall/2550amp.gif

https://elektrotanya.com/PREVIEWS/63463 ... .pdf_1.png

22k as stage 1 grid stopper gives more treble / attack than 33k and going higher than 33k can be useful.

The way you have it in your amp, a 560k divider should sound better and improve the sweep and treble. I'm speaking in terms of leaving the switch out of the equation.

As you may know, the 2n2 CC will affect loudness, fullness and push the mids in a particular way. 22n should work better in a 50W amp and will work better with more NFB. The #36 amp generally has the proper mid / vowel sound with 22n instead of 2n2. 4n7 can work well as CC because it cuts a little bass without killing the loudness or pushing the mids as much. Again, what you do in the front end of the circuit may require a change in the back end to reach your goal. The reverse of that statement is also true. Another example of this is a change in the power supply voltage having an effect on preamp voltages, thereby altering tone. Keep these relative things in mind as you tweak.

I want to stress a point I made earlier. With EL34 type tubes, 150k output grids sound rather anemic compared to 220k. If you were to put this on a switch, you would hear quite a difference. I'd go no lower than 200k there. 180k starts to trim too much for me.
:toast:

Unique
Senior Member
Posts: 229
Joined: Wed Oct 02, 2013 12:48 pm
Just the numbers in order: 13492

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by Unique » Mon Oct 29, 2018 4:33 pm

BRMSlash wrote:
Sat Oct 27, 2018 1:19 am
Been trying to clear up the bass and tame the treble, while retaining clarity. Put a 47n MTK368 on the input jack (similar to a 2555). Replaced the 1st stage cathode cap 1uf electro with a 680n MKT1813. Changed bright cap to a 2n2 1kv x5f. Put a hot shield on the 3rd stage input. Changed tone control area caps to Jupiter Reds for the bass, treble and PI input.

I'm liking the overall tone and gain, but could do with being a bit brighter I think.

http://sclk.co/s87r19

Hey, that sounds really good BMRSlash! I don't know if your going to get it sounding much better than that. I think you might have nailed it! Congrats!

BRMSlash
Senior Member
Posts: 479
Joined: Sun Jul 18, 2010 9:00 pm
Just the numbers in order: 13492
Location: Australia

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by BRMSlash » Mon Oct 29, 2018 5:31 pm

@CoffeeTones - Thanks for the tips. I'll have to try them out.

@Unique - Thanks for the kind words. I feel I've been pretty close for quite a while now, but there is a certain level of character and refinement that could do with improvement. I still feel there should be an electrolytic cathode bypass in there as well. I think the way to do it may be to setup an additional parallel 1st stage, connect the grids together and blend/mix the outputs like a standard 1959/1987.

Unique
Senior Member
Posts: 229
Joined: Wed Oct 02, 2013 12:48 pm
Just the numbers in order: 13492

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by Unique » Mon Oct 29, 2018 8:50 pm

BRMSlash wrote:
Mon Oct 29, 2018 5:31 pm
@CoffeeTones - Thanks for the tips. I'll have to try them out.

@Unique - Thanks for the kind words. I feel I've been pretty close for quite a while now, but there is a certain level of character and refinement that could do with improvement. I still feel there should be an electrolytic cathode bypass in there as well. I think the way to do it may be to setup an additional parallel 1st stage, connect the grids together and blend/mix the outputs like a standard 1959/1987.
Your welcome! One thing about it, we can only hear samples that are posted in crushed formats, through crappy computer speakers that do not have the full frequency range to hear all the details. So what we hear is limited at best. So you will know better than anyone when you have nailed the sound. I definitely think your on the right track, if not circling the wagons.

Btw, what tubes are you using for Power and preamp? I believe that tubes are every bit as important as any component in the amp. From my experience, they can have a dramatic effect on the sound, as they are an amps heart and soul. Bad sounding tubes can send you on a goose chase.

BRMSlash
Senior Member
Posts: 479
Joined: Sun Jul 18, 2010 9:00 pm
Just the numbers in order: 13492
Location: Australia

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by BRMSlash » Tue Oct 30, 2018 6:17 am

I'm using JJ ECC83MG's for the 1st three preamp tubes, JJ ECC803S's for the phase inverter and JJ 6CA7''s for the power tubes.

Unique
Senior Member
Posts: 229
Joined: Wed Oct 02, 2013 12:48 pm
Just the numbers in order: 13492

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by Unique » Tue Oct 30, 2018 9:31 am

BRMSlash wrote:
Tue Oct 30, 2018 6:17 am
I'm using JJ ECC83MG's for the 1st three preamp tubes, JJ ECC803S's for the phase inverter and JJ 6CA7''s for the power tubes.
J.J.'s preamp tubes may have some good gain, but they are some very dark sounding tubes. I've always referred to them as being 'soulless.' Changing them out will instantly bring some brightness to your amp, as well as some nice mids. I can't say much about their power tubes as I have never tried any of them. One thing about it, you can rest assure that Slash never used any J.J's during the AFD or UYI eras since they did not exist back then. All of that aside, I have to admit, even with the J.J's in your amp it sounds pretty good from what I hear.

Here's one more thing to think about when checking your amp against an actual recording where these amps are used. #36 and #34 both would sound a lot brighter in person than what you hear on any recording, mastered album or demo. This is because of the high frequency roll-off of the mics, analog recording gear, and tape being used. Not to mention the EQ'ing done by the mix engineers. Analog gear tends to roll off some high frequencies because of their natural frequency response. Analog gear also has a low frequency response cut off that rolls off some extreme lows as well, which adds to the tightness of the sound. When frequencies are rolled off, that energy spreads to other frequency ranges which boosts them a little bit. In doing so, it changes the sound a little bit.

The more analog gear that was used in the signal chain compounds the cut off frequency slopes of the various frequency responses of the analog gear used and together they become sharper. Which in turn makes the bass tighter and the highs warmer, yet still retaining a good bright sounding track. This is one of the reasons why analog recordings sounds warmer and a little tighter than digital recordings. Don't get me wrong, I'm not trying to start an analog vs. digital debate here, digital recording has come a long way in the last few years, along with engineers knowledge of how to manipulate it. However, back when AFD was recorded, AFD was the last 'all' analog recording done by Geffen Records before they started using digital processing in the signal path, and before 'Black Friday,' the merging of the big studios in the mid 90's.

So #36 in person would more than likely be brighter and not as warm sounding as what you hear in the recordings. These are things to think about when trying to get the mod correct, and your amp's sound right. Unless your trying to get your amp to sound like the album. Then none of that matters because you would probably need to change the actual mod itself from the original.

User avatar
CoffeeTones
Senior Member
Posts: 1112
Joined: Mon Oct 11, 2010 9:52 pm
Just the numbers in order: 7
Location: USA

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by CoffeeTones » Tue Oct 30, 2018 11:15 am

The made in China 12AX7 tubes most likely used in #36 are quite low gain compared to the JJ ECC83s. The medium gain, standard, EH 12AX7 tubes sound pretty good but have problems in the CF position however I've used elevated heaters to resolve that problem in some amps. Usually a Tung Sol in V1 with the rest being JJ ECC83s sounds really good in the #36 type circuit. A mix of tubes with the JJ or other tough tube in the CF position for durability is a good option. Some have gotten great tone using the old Yugoslavia made EI 12AX7 and the made in China 12AX7.
:listen:

BRMSlash
Senior Member
Posts: 479
Joined: Sun Jul 18, 2010 9:00 pm
Just the numbers in order: 13492
Location: Australia

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by BRMSlash » Tue Oct 30, 2018 5:20 pm

I agree, the standard JJ ECC83 plain & gold versions are dark and soulless. I hate them in all amp I've tried them in. However, the JJ MG's and 803S's are more open, articulate & have nicer gain. I also like the EH 12AX7's in the 1st stage of JCM2000's, however they seem too buzzy in my #36 style amp. I've also got a stash of Chinese 6N4's that I used for a while. I quite like them as well, but I prefer the MG/803 combo better. I never really like the sound of the New Sensor Tung-Sol's in any of the amps I've tried them in. I may have tried the EI's, but I can't quite recall their sound.

Unique
Senior Member
Posts: 229
Joined: Wed Oct 02, 2013 12:48 pm
Just the numbers in order: 13492

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by Unique » Tue Oct 30, 2018 7:03 pm

BRMSlash wrote:
Tue Oct 30, 2018 5:20 pm
I agree, the standard JJ ECC83 plain & gold versions are dark and soulless. I hate them in all amp I've tried them in. However, the JJ MG's and 803S's are more open, articulate & have nicer gain. I also like the EH 12AX7's in the 1st stage of JCM2000's, however they seem too buzzy in my #36 style amp. I've also got a stash of Chinese 6N4's that I used for a while. I quite like them as well, but I prefer the MG/803 combo better. I never really like the sound of the New Sensor Tung-Sol's in any of the amps I've tried them in. I may have tried the EI's, but I can't quite recall their sound.
I'm sorry, I didn't catch the "MG" designation before. I am unfamiliar with these J.J's. I guess I have brushed off J.J's for so long that I haven't paid attention to any new ones they might have came out with. I would have never guessed you had J.J.'s in your amp to begin with, so maybe the MG's are a light of hope for them.

BRMSlash
Senior Member
Posts: 479
Joined: Sun Jul 18, 2010 9:00 pm
Just the numbers in order: 13492
Location: Australia

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by BRMSlash » Sun Nov 04, 2018 3:24 am

Moved the NFB to the 8ohm tap. Changed the PPIMV resistance to a measured 220k-230k using a 1M resistor over the pot. Got rid of the cathode bypass cap and the snubber on the 3rd stage. Upped the 1st stage snubber cap to 220p. Got rid of the grid resistors on 2nd & 3rd stages. 3rd stage hot shield removed. Upped the 1st stage coupling cap to 22n. Final stage voltage divider returned to a more typical 470k//470p-470k arrangement. Increased the PI B+ node voltage.

http://sclk.co/s87ths

User avatar
CoffeeTones
Senior Member
Posts: 1112
Joined: Mon Oct 11, 2010 9:52 pm
Just the numbers in order: 7
Location: USA

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by CoffeeTones » Sun Nov 04, 2018 4:10 am

You have the best perspective since you are "in front of the amp" and you have it "under your fingers" but here is my opinion from listening. The mid voicing and articulation is terrific. The improvements I would seek are beefing up the bottom end / low mids a bit and clearing the highs a little for more clarity and thickness needed in SCOM. I would try only small things at this point because it is very good. It sounds natural, twangy, articulate and biting without ice pick. Again the mid voicing is great. It probably sounds beefier in the room than in the recording. The gain sounds right. You seem to be having fun with it. LOL

1. 110k to 120k NFB @ 8R tap
2. Larger / different value bright cap - go in steps till you find the best.
3. Possibly try the 560k thing I talked about in one or both areas of your spec.

BRMSlash
Senior Member
Posts: 479
Joined: Sun Jul 18, 2010 9:00 pm
Just the numbers in order: 13492
Location: Australia

Re: S.I.R. 100W SuperLead Schematic pt. II

Post by BRMSlash » Sun Nov 04, 2018 7:18 am

Thanks. Yeah, there's still a few things I want to change, but waiting on parts. I want to try some things with electrolytics as well.
:toast:

Post Reply