JCM800 2204 early vs later filtering config

Everything from original vintage Marshalls to reissues.

Moderator: VelvetGeorge

User avatar
stusixtysix
Senior Member
Posts: 202
Joined: Tue Aug 25, 2009 10:39 pm
Just the numbers in order: 7

JCM800 2204 early vs later filtering config

Post by stusixtysix » Sun Mar 27, 2011 6:48 pm

been looking closely at this; i'd always been under the impression that the filtering in the 2204 was unchanged with the switch to PCB-mount jacks/ pots (aka horizontal input models).

well ya learn something every day; turns out i'd never twigged that whilst the total capacitance is indeed unchanged, the later 80's models structure the filtering for the preamp stages quite differently:

early 80s 100uF main, choke, 50uF screen, 10K, 50uF PI, 10k, 50uF V2, 10k, 50uF V1. this follows the convention for 70's 50 watters, same PCB STI-202. preamp voltages from 350 V1 to 390 at PI

later 80s - 100uF main, choke, 100uF screen, 10k, 50uf PI and V2, 10k, 50uF V1. Substantially different scaling of voltages, less series resistance and a stiffer power amp.

My question, is this why the horizontal input amps have a reputation for being harsher? I wonder what Marshalls rationale was behind the change? component count is one 10k2W resistor less...

I've got one on that, now I've figured this out, I'm thinking about modding to the earlier 80's scheme, it would involve adding one resistor, removing the link from cap #2, cutting a PCB track and running one wire to the cap, creating a new seperate node for the PI and dropping the voltages to the pre lower.

As it stands It doesnt sound BAD, per se, just not that awesome either. It's been a workhorse for a long time and is due for a retube. I wonder if the earlier power supply config would be tonally benificial...?

any thoughts appreciated :)
TIA
Stu
"...i am a victim of the science age..."

gldtp99
Senior Member
Posts: 343
Joined: Wed Dec 28, 2005 11:31 pm
Location: Texas

Re: JCM800 2204 early vs later filtering config

Post by gldtp99 » Sun Mar 27, 2011 7:38 pm

i'm glad somebody else has also realized this about later horiz input 2204's------i realized this when i re-capped my 4010 1x12 JCM 800 combo in the process of modding to #34 specs/adding PPIMV----at first i thought my 4010 was different from the same era 2204's until i looked at the later 2204 schematic-----yes, they have 100uF on the output tube screens.
My 4010 was ok sounding but it was always too stiff vs other earlier 2204's and 2204-type builds---i tried many different speakers/tube swaps with no joy----then i discovered the power filtering differences when i took it out of storage to do the #34/PPIMV mods (to try and make something decent out of the amp).
I configured it to have 50uF on the screens and used a 22k in place of the last 10k dropper on the board-----i didn't want to break the trace and configure exactly as an earlier 2204 ---- in case i ever want to put this one back to stock specs---i thought that getting 50uF on the screens instead of 100uF was the most important change----- everybody loves the way this one sounds now.
Pics: http://s46.photobucket.com/albums/f112/ ... AFD%20Mod/
Note: the Sozo's and the black goop residue on the board were from the previous owner-----it was "modded" to the point of barely working when i bought it cheap several yrs ago----i had to touch up many factory solder joints on the tube sockets, poor work on this one----the Emi Wizard sounds pretty good but the amp sounds better pushing a 4x12............gldtp99

User avatar
stusixtysix
Senior Member
Posts: 202
Joined: Tue Aug 25, 2009 10:39 pm
Just the numbers in order: 7

Re: JCM800 2204 early vs later filtering config

Post by stusixtysix » Sun Mar 27, 2011 8:29 pm

gldtp99,

thanks for the info and input! A little further thought would suggest that the lower voltages in the pre would lessen headroom, give smoother preamp drive, and the lesser voltage swing outta the PI/ more PI compression from lower filtering would push the EL34's a bit less hard. This would mean less crossover notch distortion at full crank compared to the later config where the higher voltages make more headroom right through the pre and drives the power tubes harder. The 2204 I have has always had that raspy notch sound when really pushed. Was never really an issue as I always used it as a preamp-gain sound or cleanish amp anyway.

Having recently played a killer-sounding vertical input model that belongs to a friend, I think I'm gonna do this mod along with servicing the amp. This 2204 is not pristine, it had some mods (second switchable MV and mild gain boost now removed and restored to stock) so I've done a bit of work on the board before.

cheers
Stu
"...i am a victim of the science age..."

gldtp99
Senior Member
Posts: 343
Joined: Wed Dec 28, 2005 11:31 pm
Location: Texas

Re: JCM800 2204 early vs later filtering config

Post by gldtp99 » Sun Mar 27, 2011 9:00 pm

Very cool---- these horiz imput amps probably aren't ever going to be collector's favorites----especially previously modded ones---great mod platforms---go for it !!..............gldtp99

PS--- these amps can always be converted to a turret board, if needed, in case of PCB trace problems or other mods that the stock PCB won't easily support

User avatar
demonufo
Senior Member
Posts: 3882
Joined: Thu Dec 13, 2007 8:36 am
Just the numbers in order: 13492
Location: Carterton, Oxon, U.K.
Contact:

Re: JCM800 2204 early vs later filtering config

Post by demonufo » Wed Apr 06, 2011 5:07 pm

The reason that these later amps are considered harsher is actually down to slight confusion IMO. The real reason (again, my opinion) that they are slightly harsher came slightly earlier. The power transformers in the 2204 changed around '82-'83 and gave a higher B+ than before. This had quite a pronounce effect on the pre-amp side of the circuit, and gives not what I would call a harsher tone as such, but adds grit and balls, with a bit more headroom. Certainly makes for a great cleaner rhythm tone compared to the earlier circuit.

I love both however, and find it difficult to choose between the two.
So I like purple, okay!!!!!!

83.7% of all statistics are made up on the spot!

User avatar
stusixtysix
Senior Member
Posts: 202
Joined: Tue Aug 25, 2009 10:39 pm
Just the numbers in order: 7

Re: JCM800 2204 early vs later filtering config

Post by stusixtysix » Wed Apr 06, 2011 6:48 pm

demonufo, that's interesting about the PT's; my three 50 watt 800's all have B+ in the 450 - 460 region.

I've found some nice crunch tone in this amp recently after retubing with some EH 6CA7's and some NOS Philips 12ax7's and lowering the mains voltage a little bit with an autotransformer. Pushing the front with my Boss SD-1 and 80's Charvel 3A takes me right back there :D
"...i am a victim of the science age..."

User avatar
demonufo
Senior Member
Posts: 3882
Joined: Thu Dec 13, 2007 8:36 am
Just the numbers in order: 13492
Location: Carterton, Oxon, U.K.
Contact:

Re: JCM800 2204 early vs later filtering config

Post by demonufo » Wed Apr 06, 2011 7:08 pm

I know that Joey particular prefers 2204's with lower pre-amp voltages, although I can see definite benefits playing-wise to both variants. I guess the early amps are a little smoother though.
If anybody fancied trying to cop a little of that feel, you could always add another B+ dropping resistor. You may notice that there is only one in the pre-amp area of these later amps...

If you added more B+ dropping resistors at both points in the amp (the one by the pre-amp AND the one by the bias supply if required) you might find dropping the mains input voltage unnecessary. I'm not saying you will, but it's certainly an idea. And one that the filaments on your tubes might thank you for. :wink:
So I like purple, okay!!!!!!

83.7% of all statistics are made up on the spot!

User avatar
stusixtysix
Senior Member
Posts: 202
Joined: Tue Aug 25, 2009 10:39 pm
Just the numbers in order: 7

Re: JCM800 2204 early vs later filtering config

Post by stusixtysix » Wed Apr 20, 2011 10:43 pm

I was under the hood of another of my 800s yesterday and took the opportunity for a closer look at the power supply.

Amp under obs is a 4-input 50W Lead model from around 1984 and has the STI-202 JMP pcb with pots and jacks wired with flyleads. There are three 10k droppers in the B+ rail and a space on the board marked '50W link 10k 100W units'

Filter cap arrangement is 100uf main - choke - 50uf screen - 10k - 50uf PI - 10k - 50 uF v2 - 10k - 50uF v1

B+ in this amp is around 465VDC, so I'm assuming it has the later higher voltage PT as the early 80's JCM800 schematics I have show the 50 watter as having a B+ of only 365V.

Other curios - coupling cap on the bright channel is a 0.022, has a 0.0047 bright cap, NFB is 100k 4 Ohm but has 5k6 grid stoppers on the power tube sockets and 220k bias feeds indicating an EL34 model.

I'm wondering if this amp might benefit tonally from adding the extra resistor to lower the preamp and PI voltages. It's a bit raspy sounding when I crank it up and the screens really glow when pushed indicating the power tubes are being driven really hard. It's a LOUD 50 watts though, punchy, makes ya ears hurt even with my very aged 70's Greenbacks.
"...i am a victim of the science age..."

RussB
Senior Member
Posts: 256
Joined: Mon Jan 05, 2009 10:16 pm
Location: connecticut

Re: JCM800 2204 early vs later filtering config

Post by RussB » Tue Jun 11, 2013 11:56 am

stusixtysix wrote:been looking closely at this; i'd always been under the impression that the filtering in the 2204 was unchanged with the switch to PCB-mount jacks/ pots (aka horizontal input models).

well ya learn something every day; turns out i'd never twigged that whilst the total capacitance is indeed unchanged, the later 80's models structure the filtering for the preamp stages quite differently:

early 80s 100uF main, choke, 50uF screen, 10K, 50uF PI, 10k, 50uF V2, 10k, 50uF V1. this follows the convention for 70's 50 watters, same PCB STI-202. preamp voltages from 350 V1 to 390 at PI

later 80s - 100uF main, choke, 100uF screen, 10k, 50uf PI and V2, 10k, 50uF V1. Substantially different scaling of voltages, less series resistance and a stiffer power amp.

My question, is this why the horizontal input amps have a reputation for being harsher? I wonder what Marshalls rationale was behind the change? component count is one 10k2W resistor less...

I've got one on that, now I've figured this out, I'm thinking about modding to the earlier 80's scheme, it would involve adding one resistor, removing the link from cap #2, cutting a PCB track and running one wire to the cap, creating a new seperate node for the PI and dropping the voltages to the pre lower.

As it stands It doesnt sound BAD, per se, just not that awesome either. It's been a workhorse for a long time and is due for a retube. I wonder if the earlier power supply config would be tonally benificial...?

any thoughts appreciated :)
TIA
Stu


I realize this is an old thread, but I just did what stu66 was referring to. I "split" the C22 to feed the screens and PI separately, cut the trace under the yellow dot sticker right below the added 10k resistor.

100uf mains > choke > 32uf screens > 15k > 32uf PI > 10k > 50uf V2 > 10k > 50uf V1

I am experimenting with cap values at this time. C22 is now a 32/32. It significantly changes the feel of the amp. Softening the attack...it has more "give". I notice no ghosting or other anomalies. The only other mods I've made to the circuit is the removal of the C5 bright cap and replacing the 10k R29 with a 5w-15k m/o resistor. That lowers pre-amp voltages somewhat. Voltage at the PI node is ~335v


Image

Image

Image
Last edited by RussB on Wed Jun 12, 2013 2:43 pm, edited 1 time in total.

tone?
Senior Member
Posts: 250
Joined: Thu Sep 15, 2011 12:49 pm
Just the numbers in order: 13492
Location: San Francisco Bay Area

Re: JCM800 2204 early vs later filtering config

Post by tone? » Wed Jun 12, 2013 1:01 pm

So not all vertical input jcm800's have the lower b+ or all of them do?

User avatar
demonufo
Senior Member
Posts: 3882
Joined: Thu Dec 13, 2007 8:36 am
Just the numbers in order: 13492
Location: Carterton, Oxon, U.K.
Contact:

Re: JCM800 2204 early vs later filtering config

Post by demonufo » Wed Jun 12, 2013 1:08 pm

No, not all. Only the very earliest of JCM800 versions, and JMP's. Voltage/transformer change came around 82, horizontal inputs came in around April '84.
So I like purple, okay!!!!!!

83.7% of all statistics are made up on the spot!

RussB
Senior Member
Posts: 256
Joined: Mon Jan 05, 2009 10:16 pm
Location: connecticut

Re: JCM800 2204 early vs later filtering config

Post by RussB » Wed Jun 12, 2013 2:44 pm

Any comments as to what would happen if I drop the filtering to V2 and V1 to 16uf each?

User avatar
joey
Senior Member
Posts: 982
Joined: Thu Aug 20, 2009 10:40 pm
Just the numbers in order: 7
Location: Allston, Massachusetts

Re: JCM800 2204 early vs later filtering config

Post by joey » Sat Jun 15, 2013 2:03 pm

demonufo wrote:No, not all. Only the very earliest of JCM800 versions, and JMP's. Voltage/transformer change came around 82, horizontal inputs came in around April '84.
Indeed, and even some of the early -mid 70's jmp 1987's even included the additional 10k isolation resistance between the screen and PI supply nodes, in conjunction with the lower voltage drakes, which dropped the pre voltages even further to about 200vdc. The last one I worked had this.

My favorites have been the 50watt'er with the lower ht secondary drakes, not just pre voltages alone. Quite a bit changes electrically when you do this. Although I must say, the later horizontal input versions have grown on me as well! I had a 4010 come in that had the typical barky dry tone you often attribute to the latter MV models. I usually reground all Marshall's I get in for service if I'm replacing the filtration anyway, and I swear just by taking the time to more perfectly ground those amps, and to make sure it's one binding point to the chassis for the circuit reference was big and solid (often those small zinc/brass ground lugs marshall uses corrode!), the sound can improve.
Last edited by joey on Sat Jun 15, 2013 3:03 pm, edited 1 time in total.

User avatar
joey
Senior Member
Posts: 982
Joined: Thu Aug 20, 2009 10:40 pm
Just the numbers in order: 7
Location: Allston, Massachusetts

Re: JCM800 2204 early vs later filtering config

Post by joey » Sat Jun 15, 2013 2:38 pm

RussB wrote:Any comments as to what would happen if I drop the filtering to V2 and V1 to 16uf each?
That's still very good stiff filtering for a preamp and decouples at the targeted 1hz. Fact of the matter is that it really doesn't matter or buy you much for preamp stages as they present a very light and constant load to the power supply, unlike say the screens of el34's whose tug on the power supply (load) varies drasticly and suddenly with signal - in the magnitude of 10 or more times their nominal idle draw.

I would also keep filtering pretty high for the preamp too because it is more susceptiple to noise as its low level and single ended, so the cleaner the DC at that point the better.

User avatar
demonufo
Senior Member
Posts: 3882
Joined: Thu Dec 13, 2007 8:36 am
Just the numbers in order: 13492
Location: Carterton, Oxon, U.K.
Contact:

Re: JCM800 2204 early vs later filtering config

Post by demonufo » Sat Jun 15, 2013 3:16 pm

joey wrote:I usually reground all Marshall's I get in for service if I'm replacing the filtration anyway, and I swear just by taking the time to more perfectly ground those amps, and to make sure it's one binding point to the chassis for the circuit reference was big and solid (often those small zinc/brass ground lugs marshall uses corrode!), the sound can improve.
Pretty sound advice. Marshall grounding is far from ideal. I tend not to sweat it too much if they work well, but even giving the Presence a seperate ground can make a big difference. Or at the very least, seperate it from the rest of the pots, and ground it somewhere like the output jacks.
So I like purple, okay!!!!!!

83.7% of all statistics are made up on the spot!

Post Reply