Always annoyed me - that horrible hiss on those Roland JC amps...well this gent found the culprit...I did it to my JC77 and a thousand times better! A simple resistor replacement (R46 from 100k to 2.2k on the JC77...find that opamp stage on your JCXX and and do the same...)
All nice and manageable...a little volume drop, but that's welcome on this amp! Even at 2 it was way too loud......can't wait for band practice!
A possible circuit mod solution for Roland Jazz Chorus hiss... current explorations
All nice and manageable...a little volume drop, but that's welcome on this amp! Even at 2 it was way too loud......can't wait for band practice!
A possible circuit mod solution for Roland Jazz Chorus hiss... current explorations
Last edited:
Not really "noise reduction" by any means, he just cut *gain* from almost 8X to near unity gain, a huge gain loss.
More than "a little volume drop" , some huge 18dB
If you record that amp, S/N will be exact same as before, because you will need to raise mixer gain by exact same amount.
Idea originator comments:
Of course, the proper solution is to replace old JRC4558 with a modern low noise one.
An adapter board will be needed because old ones are SIP8 while modern replacements typically come in SMT packages, but some entrepreneur will certainly offer something like that. (hint hint).
MOD as shown might be useful only when bedroom playing, in a very silent environment, but not in any kind of "live" playing.
FWIW here is relevant schematic:
More than "a little volume drop" , some huge 18dB
If you record that amp, S/N will be exact same as before, because you will need to raise mixer gain by exact same amount.
Idea originator comments:
No, it´s not, it´s exactly what´s needed to recover from passive EQ losses.
I bet there will be much eye-rolling in Japan
Of course, the proper solution is to replace old JRC4558 with a modern low noise one.
An adapter board will be needed because old ones are SIP8 while modern replacements typically come in SMT packages, but some entrepreneur will certainly offer something like that. (hint hint).
MOD as shown might be useful only when bedroom playing, in a very silent environment, but not in any kind of "live" playing.
FWIW here is relevant schematic:
Sorry but no.
Neither Signal reading or schematic are wrong, but you are seeing half the picture
1) the Op Amp stage gain is (100/15)+1=7.6666X , round it up to 7.7X or straight 8X (nearest round value).
Mod lowers that to (2.2/15)+1=1.14X ... practically 1X
2) BUT you are ignoring the passive tone control stack loss, which is huge.
Notice the original signal level of 135 mV is before it.
In Jones type EQ (often called Baxandall, which they are not) normal loss is 10X or 20 dB so typical recovery stage is 10X gain.
In simpler "boost only" Fender type ones, which Roland uses, loss is somewhat less, about 7X , so that is the proper gain set for the next "recovery" stage.
And that´s what we find here.
Again, that "Mod" might be acceptable at home, but definitely makes the amp almost unusable Live, at a Club or even a Church, which is a relatively silent place but large and full of people.
hi guys,
in my humble experience, a simple replacement of the "hissy" opamps with not "hi fi" jfet based ones, but with old-school "low fi" ones like 1458 can make a HUGE difference in the hiss issues in solid state amps.
may seem counterintuitive, but hear me out...
tl0x2 style chips have jfet inputs... jfets DO sound like... "tubish" to some people when run clean... but when run full up, tend to be much hissier with a lot more high end... and when they clip, they clip hard and ragged, which can make the hiss worse.
a bjt front end will still clip, of course, but it doesn't have anywhere near the frequency response... it simply can't reproduce the high end that the tl0xx's generate.
that's why the 1458's were used in the acclaimed solid state marshalls. tho they may be "inferior" on paper, in application, they end up being superior in this respect.
recently i aquired another marshall 5212 in a hoss trade.
very clean, sounded great, but when cranked up, it hissed so bad i had to run a noise gate, which was a real pain in the tail to deal with, and adversely affected the use of the amp.
i mean the guys in the band were bitching. and hard. the amp really sounds best with the preamp gain pegged, and then it was so noisy so as to be unuseable.
so anyways, last week, i opened it up planning to do a cap job and run thru the preamp to try and find the source of the noise.
to my surprise, someone had removed all the original jellybeans and replaced them with socketed tl072's. i personally don't care for the way they sound when clipped at all... very edgy and spikey.
first things first, i removed all of them, and replaced them with the stock 1458's in the amp as in my others.
almost every bit of the hiss went away instantly.
i can now crank the amp literally full blast, and no more noise than any other amp cranked up, and no more hiss. guitar sounded warmer and was more responsive too.
i don't know if this will help with a jc, as i recently sold mine, but i humbly suggest if replacing opamps, you try not only hi-fi chips, but lower fi ones... 1458, 4558, etc... it may not only improve your noise floor <cuz it can't produce the frequencies high enough to really amplify the hiss> but may give you a better overall tone by shifting the amplified frequencies down into a more guitar friendly range where the amp will overall function better.
i am no ee. just a monkey with a breadboard and a willingness to f around and find out, and this was what i found recently, and it seemed it may be relevent to the issue being faced.
submitted in hope this may help someone. i may be wrong, and it may not apply in all cases, but it definitely works this way with the distortion pedals i work with for a living as well.
in my humble experience, a simple replacement of the "hissy" opamps with not "hi fi" jfet based ones, but with old-school "low fi" ones like 1458 can make a HUGE difference in the hiss issues in solid state amps.
may seem counterintuitive, but hear me out...
tl0x2 style chips have jfet inputs... jfets DO sound like... "tubish" to some people when run clean... but when run full up, tend to be much hissier with a lot more high end... and when they clip, they clip hard and ragged, which can make the hiss worse.
a bjt front end will still clip, of course, but it doesn't have anywhere near the frequency response... it simply can't reproduce the high end that the tl0xx's generate.
that's why the 1458's were used in the acclaimed solid state marshalls. tho they may be "inferior" on paper, in application, they end up being superior in this respect.
recently i aquired another marshall 5212 in a hoss trade.
very clean, sounded great, but when cranked up, it hissed so bad i had to run a noise gate, which was a real pain in the tail to deal with, and adversely affected the use of the amp.
i mean the guys in the band were bitching. and hard. the amp really sounds best with the preamp gain pegged, and then it was so noisy so as to be unuseable.
so anyways, last week, i opened it up planning to do a cap job and run thru the preamp to try and find the source of the noise.
to my surprise, someone had removed all the original jellybeans and replaced them with socketed tl072's. i personally don't care for the way they sound when clipped at all... very edgy and spikey.
first things first, i removed all of them, and replaced them with the stock 1458's in the amp as in my others.
almost every bit of the hiss went away instantly.
i can now crank the amp literally full blast, and no more noise than any other amp cranked up, and no more hiss. guitar sounded warmer and was more responsive too.
i don't know if this will help with a jc, as i recently sold mine, but i humbly suggest if replacing opamps, you try not only hi-fi chips, but lower fi ones... 1458, 4558, etc... it may not only improve your noise floor <cuz it can't produce the frequencies high enough to really amplify the hiss> but may give you a better overall tone by shifting the amplified frequencies down into a more guitar friendly range where the amp will overall function better.
i am no ee. just a monkey with a breadboard and a willingness to f around and find out, and this was what i found recently, and it seemed it may be relevent to the issue being faced.
submitted in hope this may help someone. i may be wrong, and it may not apply in all cases, but it definitely works this way with the distortion pedals i work with for a living as well.
I found 4558 to be dreadfully hissy in 50x gain in a RIAA phono input stage. This was in a disco mixer.
The reason pinkjimiphoton is probably enamored with 1458 is that in guitar & bass amps, the speaker does high frequency cut. Also many US males are deaf to frequencies over 7 khz. My hearing goes to 14 khz.
I did a subtitution of 4558 with low noise 33078 in my RIAA circuit, and found they oscillated ~ 1 mhz. Solution, a 33 pf disc cap across the feedback resistor and a .1 uf ceramic cap between +- rails 1" from the two op amp packages. The Jazz Chorus in post 1 already has 150 pf across the feedback resistor, so the main problem with 33078 is DIP package instead of SIP. NJM2068 has noise specs similar to 33078 and comes in SIP package. My only source NJR parts is Mouser who charges $15 freight always so I don't deal with them.
The reason pinkjimiphoton is probably enamored with 1458 is that in guitar & bass amps, the speaker does high frequency cut. Also many US males are deaf to frequencies over 7 khz. My hearing goes to 14 khz.
I did a subtitution of 4558 with low noise 33078 in my RIAA circuit, and found they oscillated ~ 1 mhz. Solution, a 33 pf disc cap across the feedback resistor and a .1 uf ceramic cap between +- rails 1" from the two op amp packages. The Jazz Chorus in post 1 already has 150 pf across the feedback resistor, so the main problem with 33078 is DIP package instead of SIP. NJM2068 has noise specs similar to 33078 and comes in SIP package. My only source NJR parts is Mouser who charges $15 freight always so I don't deal with them.
Last edited:
my hearing is tested good to 16 k still, so nope.
also, the speakers i use aren't the stock celestions. so, nope. i use jbl's or peavey scorpions, which .... if you think about it, as they produce more highs than a celestion.... would exacerbate hiss, not roll it off. if the amp was hissing with the jfet jelly beans, and are replaced with the bjt jelly beans and the hissing increased or stayed the same, you MAY have a point. but it doesn't. just the opposite, in fact.
there's a commonly held illusion that jfets sound better. its internet bunk and "mojo" and despite "stats", in reality, it doesn't always work out that way. i've auditioned hundreds of chips looking for the ones that work best with guitar signals, and the one that always gets chosen by beta testers of the circuits have one chip in particular they always choose.
i've encountered this phenomenon on several "broken" amps i've picked up cheap and repaired.
you gotta look at the frequency response of the chip to actually understand it. a 1458 can't produce the frequencies of the hiss. particularly when driven hard. its the chip that rolls off the high frequencies, not the speakers, tho speakers can indeed have that effect in part.
oscillation around 1 mhz would be way too high for ANY human to hear. it wouldn't likely manifest as hiss, but distortion imho by ring modulating any signal being put thru the circuit.
my point is that automatic assumption that a "hifi" chip will improve hiss in a guitar amp is not always correct.
the amp can't amplify those frequencies if they're rolled off by the circuit before it ever hits the power amp.
peace.
also, the speakers i use aren't the stock celestions. so, nope. i use jbl's or peavey scorpions, which .... if you think about it, as they produce more highs than a celestion.... would exacerbate hiss, not roll it off. if the amp was hissing with the jfet jelly beans, and are replaced with the bjt jelly beans and the hissing increased or stayed the same, you MAY have a point. but it doesn't. just the opposite, in fact.
there's a commonly held illusion that jfets sound better. its internet bunk and "mojo" and despite "stats", in reality, it doesn't always work out that way. i've auditioned hundreds of chips looking for the ones that work best with guitar signals, and the one that always gets chosen by beta testers of the circuits have one chip in particular they always choose.
i've encountered this phenomenon on several "broken" amps i've picked up cheap and repaired.
you gotta look at the frequency response of the chip to actually understand it. a 1458 can't produce the frequencies of the hiss. particularly when driven hard. its the chip that rolls off the high frequencies, not the speakers, tho speakers can indeed have that effect in part.
oscillation around 1 mhz would be way too high for ANY human to hear. it wouldn't likely manifest as hiss, but distortion imho by ring modulating any signal being put thru the circuit.
my point is that automatic assumption that a "hifi" chip will improve hiss in a guitar amp is not always correct.
the amp can't amplify those frequencies if they're rolled off by the circuit before it ever hits the power amp.
peace.
No,that´s an average loss when pots are set "flat".
Roughly you lose 7X, gain 7.7X so 135mV before tone control and gain stage become indicated 150mV with tone controls flat.
Of course rising the to max ("diming") will strongly increase lows and highs, it´s their job!
Surprisingly having very little or no effect at 1kHz or whatever turnover frequency was used, because in general tone controls have a central frequency at which they "do nothing".
Load those values in TSC: Duncan´s Tone Stack designer and move (virtual) sliders yourself.
TSC
You´ll understand everything.
Including passive tone control loss.
These are approximately "James" type.
Here is one explanation, 2 details:
1) notice how it *starts* with 20dB attenuation (10X) when set flat
2) it "does nothing at about 700Hz
3) only problem is they WRONGLY call this a "Baxandall" tone control, they even mention the original article ... which they obviously never ever read .
Proper name is James.
Roughly you lose 7X, gain 7.7X so 135mV before tone control and gain stage become indicated 150mV with tone controls flat.
Of course rising the to max ("diming") will strongly increase lows and highs, it´s their job!
Surprisingly having very little or no effect at 1kHz or whatever turnover frequency was used, because in general tone controls have a central frequency at which they "do nothing".
Load those values in TSC: Duncan´s Tone Stack designer and move (virtual) sliders yourself.
TSC
You´ll understand everything.
Including passive tone control loss.
These are approximately "James" type.
Here is one explanation, 2 details:
1) notice how it *starts* with 20dB attenuation (10X) when set flat
2) it "does nothing at about 700Hz
3) only problem is they WRONGLY call this a "Baxandall" tone control, they even mention the original article ... which they obviously never ever read .
or they have a GROSS lack of text interpretation.
Proper name is James.
Don´t think hiss will be an isue when playing with a band, at all.
Those are GOOD amps, well designed.
Hiss only becomes annoying at home, in a very low background noise environment.
That said, it´s worth improving Op Amps, I guess there is an adapter board which lets you use modern cleaner SMD Op Amps
You will still have some hiss, it´s a Physical reality, but less of it.
Those are GOOD amps, well designed.
Hiss only becomes annoying at home, in a very low background noise environment.
That said, it´s worth improving Op Amps, I guess there is an adapter board which lets you use modern cleaner SMD Op Amps
You will still have some hiss, it´s a Physical reality, but less of it.
SO/MSOP/TSSOP/SOIC/SOP8 to DIP8 PCB Board
– QuartzComponents
Buy online 8PIN SMD TO DIP Adapter in India at low cost from DNA Technology.
Cheap...
– QuartzComponents
Buy online 8PIN SMD TO DIP Adapter in India at low cost from DNA Technology.
Cheap...
Last edited: