een7gdr

Hi,

I have noticed that with the effects loop, the amp sounds much thinner than without the loop connected. I have linked the loop straight through and tried both +4/-10 settings which both sound similar.

Im using the amp at low bedroom levels and can still hear sound coming though the amp on the red channel with the level at 0 when using the loop. This sound is very tinny and i think is the source of the problem - the tinny signal adding to the normal but quiet signal = thin sound. I believe that at normal volumes (4+) the actual signal would swamp the tinny signal so much you couldnt notice.

Can people try linking out their loops (DSL and TSL as they share similar loop circuits apart from serial/parallel) and let me know if you can still hear the red channel at 0 volume settings? I have read that voodoo do an fx loop mod that is supposed to make the loop more transparent but i dont fancy shipping the amp to the states for a $25 mod!

I am an electronic engineer and will be doing to playing with the circuit, however i dont have much experience with amps (yet!).

00jett

im interested in your progress as i have the same issue with my dsl. i also looked at the voodoo mod but just dont have the cash

een7gdr

Well,

Ive been scoping the FX return circuit and there is a constant AC audio signal superimposed onto the +/- 15V power lines to the FX return amplifier IC. This signal remains a constant amplitude no matter what volume level is set.

I have tried changing the IC from a gerneral purpose RC4558 to a few other hifi quality op-amps and whilst the noise level may have reduced a little on the fx loop the tinny signal at 0 volume persisted.

Ill just have to keep looking for now!

Incidentaly, the voodoo FX mod is only $25 so must be a pretty simple fix. Whilst it is supposed to make the loop more transparent i dont know if it still possess the same 0 volume "feature".

00jett

very interesting. i have noticed what your saying about the noise being unaffected by volume. if you find something that works post i. please lol. you have got my attention. i use rack gear some times so that noise really bothers me.

zeusecho

I have the same issue, just waiting to get some cash and have the voodoo effects loop mod along with some others while they have it.

__________________
Marshall DSL50
H&K Switchblade 50W Combo
Line6 4x12 (Hellatone 30s and 60s X pattern)

Gibson LP Custom
ESP MH-301
Merz Custom Acoustic
Taylor 355ce 12 String

BrotherJ

Tried mine on 0 & no effects loop sound came through at all. Still i'm not 100% happy with the loop on my TSL, i'd be interested in seeing if it can be modified to work better...

jcmjmp

I've never has this problem or even heard about it before. Also, the TSL loop is not the same as the DSL so you might want to concentrate on one amp or the other.

een7gdr

There are plenty of people that complain about the effects loops, me included. Im just trying to narrow down where the change in tone is coming from.

People often blame the problem on it using ICs rather than pure tubes throughout the signal chain however the fx loop is not supposed to colour the sound, so apart from introducing noise through attenutation / re-amplification a good design should not produce such noticable effects. After all i cant see why tube fx loop buffers are any better when the effects themselves will almost certainly use ICs.

The TSL loop is different overall however the first IC post fx loop is identical, albeit with a few R/C changes. This is the one that amplifies the returned signal before the mixing process. If both amps exibited the same problem then it might narrow down the search.

I should have asked people to specify which amp they tested. jcmjmp you say youve never come across this issue, do you have a dsl or tsl to try it on?

I had planned to do some trials this weekend however the components i ordered didnt turn up today so itll take a little longer.

zeusecho

Check the sig, my problem child is a DSL50. If you get a fix I'd love to know.

__________________
Marshall DSL50
H&K Switchblade 50W Combo
Line6 4x12 (Hellatone 30s and 60s X pattern)

Gibson LP Custom
ESP MH-301
Merz Custom Acoustic
Taylor 355ce 12 String

JeremyallenGTR

Mine does the same thing. The effects loop on the DSL is weak and needs some work!

een7gdr

Well,

After some tweaks the mod ive been working on appears to improve things, see rather rushed veroboard prototype below!

The problem is due to the DSL 50/100s sharing an IC to buffer the FX send signal (which can be at a low level depending on the volume) and some of the tone circuits. The tone circuits are at a very high voltage level and are inducing some of the signal into the FX send circuit. The mod below moves the FX send buffer to a seperate IC and shows heavily reduced crosstalk.


It plugs in between two connectors, CON10 on the JCM2-61-00 schematic (tone control PCB board), and CON3 on the JCM2-62-00 schematic (Rear PCB with FX loop jacks). One wire needs to be soldered, i have taken this to the common pin on RL2b that connects to the top of R40.

It pretty much replicates the circuit around IC2a just after the volume pots and RL2b that buffers the controlled signal before hitting the power stages. The new buffer inserts its signal into pin 1 of CON10 where the old signal went so no further changes are neccesary. As i am using the new circuit seperate from the tone board, the cross-talk is reduced dramtically so you can hardly hear it on 0 volume when the loop is used. Since the signal is so small i can now play on normal bedroom volume levels with the loop without affecting the tone any where near as much as the standard DSL, and the effect of the loop does not now vary with volume.

I had to be very careful as to minimise track wires between the high impedance attentuator (R40, C26 and R39) and the new IC2a circuit. This is where the original circuit is suscepable to the interferance.

The veroboard circuit is a little messy, it really needs a proper PCB making to ensure crosstalk is reduced further. Ill be looking into this next.

Attached Images

fx_test_small.jpg (89.7 KB, 47 views)

SaveTheMarshall

Fantastic work. I will be checking back to see your final progress. I hate my FX loop and would love to replicate this.

solarburnDSL50

__________________
DSL50 - FJA modded
Loaded with a Martimus Maximus Preamp Tube Stack
Power tubes 6CA7EH's

Strat HSS & Tele HH

Avatar 412 loaded with Greenbacks

Vox Night Train
Russian NOS 84's
Telefunken & RFT ECC83

BrotherJ

Hey looks good, cheers for posting it!!

You mentioned that you replaced the RC4558 with a new chip & it reduced the noise a bit, I have a TI RC4558 that I was planning on installing - is it worth the trouble? My main issue was a bit of noise when the loop is on the +4 setting.

__________________
TSL100 - MM OT & Choke and some circuit mods - 1960AV
Stiletto Deuce II - Traditional quad

een7gdr

My amp had a TI RC4558 in it from the factory, it was built in 1998. There are numerous compatible chips on the market - the RC4558 is a fairly cheap general purpose amp. Burr Brown do a OPA2134PA version that is more hi fi quality. The reduction in noise i mentioned was measured on a scope, the reduction was measureable but i wouldnt say noticable and not worth the effort of actually removing the chip from the board.

It seems strange that you mention you get noise on the +4 setting, this is the least noisy option in my opinion! The FX loop works by reducing the level of the original signal to levels compatible with either pedals or rack effects. After the effects, the levels are then boosted back up to the original levels. The +4 level is highest signal level of the 2 options, hence the original signal is not reduced as much and there is less amplification after the effects loop. This should mean there is a little less noise that on the -10dB option.


I have ordered 30 PCBs for the final version, the smallest number that made it worth ordering. If anyone is interested i can sell the completed kit,fully populated PCB, extra patch cable and mounting posts for around £17 plus any extra postage to cover overseas shipping. Once i have the boards ill post images etc.

SaveTheMarshall

Sign me up. I'll take one.

MaidenStrat02

Same here, I want one too. I'm DSL50SG07Abuser on "The Marshall Amp Forum" also. I've been watching and replying there also just to let you know. How do you plan on selling them? Do you have a PayPal account? That would be easy for alot of people I would imagine.

__________________
Gibson '07 SG
2 Customized Fender Strats
Modded DSL50 MM choke,PT&OT
Carvin 2x12 V30's
Pitchblack-LPB-1-Budda Wah-TS808. Loop= 108EQ-Modded Phase 90-Carbon Copy-George L cables-Voodoo Pedal Power 2 Plus-Pedaltrain board

Contact for Facebook request

00jett

interested

een7gdr

What ill do is take all the names that have requested them, up to 29 people (i want to keep 1 circuit back for me!). Ill pm them all with my paypal address and send the units out once payment has been received.

Itll be at least another week before the PCBs get to me, but as soon as they do ill post pics etc of the whole thing. Those you have stated their interest already will be first on the list.

If i get more than 30 or so people ill have to buy in some more PCBs but id be grateful of feedback from those giving it a go.

Incidentally, when most people complain of tone suck i think it the pedals to blame. I have improved the sound of my amp, 100% at low volumes through the FX loop but the more pedals i add to the loop, the more "difference" you can hear when you activate the loop. I just cant believe that Marshall let the loop go with the appalling tinnyness we were experiencing! Maybe this is why marshall didnt add an FX loop switch - because it made the sound more noticable! I also think this is why some people sell/buy attenuators from ebay for a fiver that just act as a basic volume control in the fx loop. It allows them to turn up the proper volume and try to overcome the cross-talk issue hence the amp sounds more like it should without the loop - albeit with a lot of extra noise/hiss!

00jett

easy, detailed instructions will help too...

SaveTheMarshall

When you get around to contacting us I'm also SaveTheMarshall over at marshallampforum. Just letting you know since there is a thread over there as well. So you can PM me either place. Thanx.

een7gdr

No problem guys, will try my best to make it even between the 2 forums.

Ill print out insturctions etc and the schematics so you know whats going on. There will only be 1 wire to solder onto the board. In fact i used a croc clip for initial testing onto a resistor leg. Ill print it all out anyhow. Luckily, most signals i needed were already on one of the white connectors, hence i use a patch cable and insert my board in the middle of the cable. Ive been told the PCBs will be posted to me on the 11th.

Currently I have the following interested:

1) SaveTheMarshall
2) MaidenStrat02
3) hotguit (other forum)
4) 00jett

Ill most likely put a socket on the IC so that people can play/swap out different types if they want to.

Cheers,
Graham

solarburnDSL50

I was playing my amp today and had a pedal in the loop. I got the volume up to 4/5 and it started to thin out. I turned the pedal off in the loop and noticed the tone come back some. I thought its the loop. I unhooked cables and pedal out of the loop and bam all the tone was there at volume no thinning out.

Found my thinning out problem at higher volume. Most of the time I'm playing around 2 on the volume and you don't notice it until about 4 and up. Right now I'm going to see what Jerry from FJA recommends doing. I really depend on the loop for delay and a aural xciter at least.

__________________
DSL50 - FJA modded
Loaded with a Martimus Maximus Preamp Tube Stack
Power tubes 6CA7EH's

Strat HSS & Tele HH

Avatar 412 loaded with Greenbacks

Vox Night Train
Russian NOS 84's
Telefunken & RFT ECC83

BrotherJ

Cheers for the reply, It's an odd thing but it does have a bit more noise than the -10 setting for some reason. I'll check it out.

With your kit, you mention you are by-passing the IC2a circuit off the preamp board and using your own circuit to reduce the noise, I was curious if it helps cure the sound thinning out at higher volumes??

__________________
TSL100 - MM OT & Choke and some circuit mods - 1960AV
Stiletto Deuce II - Traditional quad

BrotherJ

By the way, Cheers for posting this mod!!

__________________
TSL100 - MM OT & Choke and some circuit mods - 1960AV
Stiletto Deuce II - Traditional quad

MaidenStrat02

Great, can't wait to hear from you. Thanks for figuring this out. I know alot of users, myself included, are going to love this mod. Thanks again, Ron.

__________________
Gibson '07 SG
2 Customized Fender Strats
Modded DSL50 MM choke,PT&OT
Carvin 2x12 V30's
Pitchblack-LPB-1-Budda Wah-TS808. Loop= 108EQ-Modded Phase 90-Carbon Copy-George L cables-Voodoo Pedal Power 2 Plus-Pedaltrain board

Contact for Facebook request

een7gdr

As solarburn mentions, the loop on its own is not at all bad at higher volumes. I believe the problems seem to be with the pedals themselves. On a higher volume the signal going to the pedals is greater. Most amplifier circuits are good/linear at low volumes but start to go non-linear at higher signal levels. Pedals with a normal off signal path (those without a true bypass) tend to use a simple transistor buffer circuit - these work great at low signals but will introduce distortion and affect frequency response at higher levels. Components such as transistors and diodes have their own inherent capacitance - this tends to vary with voltage so it is also possible that the larger signals change the parasitic capacitance of the components and change the freqency response a little normally this is only slight but depending on the circuits may be noticable. My major issue is the amount the thinning changes with volume at the low end - this just cannot be dialled out on the eq.

I noticed that the loop, before and after my mod, seems to drop a little volume when going through the pedals - the more pedals i add the greater the drop. I am pretty sure this will be the case with any serial effects loop as the drop is happening in the pedals. I find it a little difficult to judge differences between the pedals on/off tone suck as the pedal obviously affects the signal when on.
On another note, this is why i would not want a true bypass in the "loop" pedals - the on/off circuits both affect tone a little however they often share similar characteristics so that the on/off modes are reasonably well matched. Overall effects of these can then be dialled out on the amp or eq. If you suddenly switch to true bypass pedals you would get sudden changes in tone/volume when switching pedals out. For volume/gain pedals a true bypass is ideal since you would expect a change in signal level betweem the on and off modes.

If any of the tone suck at higher volumes was due to the amp it should still be there when the loop is linked out.

een7gdr

Right then, the boards are in. The mod consists of the following:

PCB with isolated DSL Effects Buffer Circuit
Short Patch Cable
A couple of PCB adhesive mounts

The trailing wire needs to be soldered onto the relay situated between the reverb channel A pot and the red channel volume pot - this is the only solder joint needed - the board and patch cable simply plug in to the connectors already on the board. Once connected it can be mounted either onto the PCB nearby or onto the metal fascia - See images below. Im happy with the mounting in my rig - if your worried about the unit falling off you can wrap it in insulation tape so that it wont short to anything. My first prototype just hung from the patch cables and was wrapped in black tape just in case!

As i have mentioned, this prevents the effects loop thinning out at bedroom levels due to high frequency crosstalk between the buffer chip used on the stock models. The crosstalk is a constant level regardless of the volume so as you turn up the volume, the sound starts to sound more like it should. The crosstalk is always there, you just swamp it out as you turn the volume up. This makes the thinning very dependant on volume and was damn annoying for me! This mod removes the crosstalk as it uses an isolated buffer chip that has no other signals running through it and hence keeps the effects loop transparent all the way down to 0 volume.

If you only ever use the amp at high volume levels you may never notice the crosstalk and most likely wont need the mod. Personally once i know the problem is there, it just bugs me!

Ill PM those that have expressed an interest with regards to purchasing the kit.

The Kit:

Mounting Option 1:

Mounting Option 2:

Solder Position

Solder Position

SaveTheMarshall

Beautiful work.

__________________
JCM2000 DSL50
Mercury Magnetics Output Transformer
Mercury Magnetics Power Transformer
Mercury Magnetics Choke


http://www.marshallampforum.com/foru...pic.php?t=7917

MaidenStrat02

Here's an update everyone. You have to buy this Mod, no joke this is the real deal. It was super easy to install, the instructions and pictures made it simple for people with basic knowledge of amp repair. Just make sure there is no stored Voltage in the main Caps and you're fine. Now we get to what makes this so special. Remember how you could hear your guitar with the volume at zero on either channel and your guitar volume turned up? NOT ANYMORE!! I had to bend down and put my ear to my cab to hear any bleed through. No more hiss or leakage of signal is pretty darn sweet. The mod translates out to be $33 American because Graham is in the U.K. He is great to talk to and was right there if I needed anything.

__________________
Gibson '07 SG
2 Customized Fender Strats
Modded DSL50 MM choke,PT&OT
Carvin 2x12 V30's
Pitchblack-LPB-1-Budda Wah-TS808. Loop= 108EQ-Modded Phase 90-Carbon Copy-George L cables-Voodoo Pedal Power 2 Plus-Pedaltrain board

Contact for Facebook request