Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: PharmRock on August 24, 2021, 05:45:01 pm
-
While doing some research for a JTM45 build I am putting together with the idea of using a switchable SS vs tube rectifier, I came across this thread https://el34world.com/Forum/index.php?topic=26441.0 (https://el34world.com/Forum/index.php?topic=26441.0) that discusses the Dr Z EMS amplifier, which is a JTM45-ish amp that happens to have switchable SS/tube rectification.
Also on this amp is a "touch control", which seems (to the best I can tell) inserts a 0.047uF cap from the first filter stage/OT node to ground (see pic). I am trying to piece together what else is going on just based on what I can see, but honestly can't tell. And...as many of you know...I'm still in the pre-K stages of learning this stuff. I've attached an image that highlights A) what I know is the "touch" switch and B) what I think it is switching in/out.
Can any of you experts offer up an explanation for what this is, as well as what the effect would be on the tone (i.e. what is the "touch" referring to)?
Thanks!
-
Post a schematic if you want more than just speculation.
-
Post a schematic if you want more than just speculation.
Unfortunately, I can't seem to locate one. I'll keep looking.
-
I doubt you’ll find a schematic. I’m learning too but I suspect the switch activates the second big filter cap that you see there, effectively doubling or halving the first filter stage capacitance. The .047 cap may just be there to prevent popping. :dontknow:
Doubling the capacitance would firm up the feel or “touch”. Halving it would be a looser and more open feel.
-
I’m learning too but I suspect the switch activates the second big filter cap that you see there, effectively doubling or halving the first filter stage capacitance.
Those 2 big caps are in series, their only 300v for 600v total, so their not using a single cap.
-
It could also be there as a bypass around the bigger caps (common in Hi-Fi stuff)?
-
Looks like the two big electrolytic caps are in series and permanently in the circuit. They do not seem to be affected by the switch.
The key to the switch function is the brown wire connected to the right side of the yellow cap. This is connected to ground when the switch is closed. When the switch is open, the cap is in series with the brown wire.
If you can find out where the brown wire goes, we may be able to figure it out.
EDIT: hmm… on second thoughts, it just seems the switch connects the yellow cap in parallell with the filter caps. That should make very little difference. I think there is something else we can’t see here…
-
Hi,
Premier Guitar mag did a review quite some time ago.
The .047uF I think bypassed the bottom ?? Filter cap ??
I will see if I can find the review.
John
-
Hi,
Guitar player.
https://www.dropbox.com/s/hte66ykpa3ammuf/EMS%20BB212%20Guitar%20Player%20ReviewAUG%2017.pdf?dl=0 (https://www.dropbox.com/s/hte66ykpa3ammuf/EMS%20BB212%20Guitar%20Player%20ReviewAUG%2017.pdf?dl=0)
John
-
Hmm... having a small cap between the center tap and ground doesn't make sense to me (or does it?). I'm thinking it's a misunderstanding by the journalist. It doesn't match what I interpret from the picture either (I could be wrong of course!).
I love these types of investigations. It's a fun challenge!
-
That review states that the touch switch just connects a 47nF cap across the HT, but it would be better to verify the circuit. There’s 2 red wires and a brown wire connected to that node, which need to be tracked and identified.
-
Theory:
The two 220u caps are the first filter node. If they have high ESR (internal resistance) caps, the response to peaks could be a bit sluggish or saggy. Switching in the 47n cap could get a faster peak response.
Found some data sheets for similar IC caps, listing EC of about 1.5 ohms (or about 3 ohms, since it's two in series). Not sure if 3 ohms would make a noticeable difference though.
Just a wild theory... feel free to shoot it down! :)
-
I think Dr Z is known for doing some circuit changes involving the OT primaries, such as a conjunctive (or corrective) filter as seen on his Carmen Ghia amp (borrowed from the Watkins Dominator). But IIRC there is usually a resistor in series with the cap. Not saying this is what we're looking at but maybe along the same lines.
The brown wire appears to begin from the rectifier switch, which is adjacent to the impedance selector. It (together with a black and red wire in a twisted bundle) follows the back of the chassis and then along the right side and comes up under the board twisted together with a black (or maybe blue) wire by the right-most edge of the board. Perhaps the red wire pair is the OT center tap and one of the choke wires?
-
Theory:
The two 220u caps are the first filter node. If they have high ESR (internal resistance) caps, the response to peaks could be a bit sluggish or saggy. Switching in the 47n cap could get a faster peak response.
Found some data sheets for similar IC caps, listing EC of about 1.5 ohms (or about 3 ohms, since it's two in series). Not sure if 3 ohms would make a noticeable difference though.
Just a wild theory... feel free to shoot it down! :)
Kudos on checking the ESR spec!
I’m not sure it’s a go’er though, eg otherwise 1ohm cathode current sensing resistors would have a rep as tone suckers.
It may be notable that the review didn’t seem to mention the touch switch.
-
From the manual:
"Attenuates the transient response and pick attack. Set to down position to attenuate attack or for bright single coils. Set to up position for fast transient response and cut"
TRANSIENT RESPONSE Electronics Assignment Help and Homework Help - TRANSIENT RESPONSE Project Help (electronicsassignments.com) (https://www.electronicsassignments.com/transient-response-12399)
-
Theory:
The two 220u caps are the first filter node. If they have high ESR (internal resistance) caps, the response to peaks could be a bit sluggish or saggy. Switching in the 47n cap could get a faster peak response.
Found some data sheets for similar IC caps, listing EC of about 1.5 ohms (or about 3 ohms, since it's two in series). Not sure if 3 ohms would make a noticeable difference though.
Just a wild theory... feel free to shoot it down! :)
Kudos on checking the ESR spec!
I’m not sure it’s a go’er though, eg otherwise 1ohm cathode current sensing resistors would have a rep as tone suckers.
It may be notable that the review didn’t seem to mention the touch switch.
I did check ESR for a few other cap brands and they seemed to have somewhat lower ESR (like 0.5 ohms) for similar caps. Still seems unlikely that it would make a significant difference though.
-
So it’s like a dull cap on the first B+ node, similar in concept to what you would see on V1 on a Friedman or higher gain marshall style amp? Usually a 500pf from plate to ground.
-
So it’s like a dull cap on the first B+ node, similar in concept to what you would see on V1 on a Friedman or higher gain marshall style amp? Usually a 500pf from plate to ground.
Electrolytic caps distort on the frequency extreme ends and are slow in response on the high frequencies compared to film caps.
So they put a small film cap in parallel with the lytic filter cap to clean up the highs and speed up the high end response.
Hi-Fi guys do this. You used to be able to only get so large a value film cap, now you can get large value film caps, but their very large in size too. Any where from, say, 0.47uF to 1uF should work. The bigger the better within reason. Some put it across the 1st B+ filter cap, others put it across the last (preamp tube) filter cap.
-
Hi,
Looks as if is across the first PSU node to grnd when switched.
Mesa Boogie do this, I always thought it was to squelch any arching when activating the STBY switch ??
While the sluthing hat is on, can anyone tell me why the green cap across V2 cathode resistor(820 ohm) is smaller than its brother across V1 2k7 cathode resistor, it was reported as both being 0.68uF.
Different voltage ? smaller value ??....maybe.
May be I need to get a full time job !!
John
-
Hi,
Looks as if is across the first PSU node to grnd when switched.
Mesa Boogie do this, I always thought it was to squelch any arching when activating the STBY switch ??
While the sluthing hat is on, can anyone tell me why the green cap across V2 cathode resistor(820 ohm) is smaller than its brother across V1 2k7 cathode resistor, it was reported as both being 0.68uF.
Different voltage ? smaller value ??....maybe.
May be I need to get a full time job !!
John
Most likely, one of the caps are not 0.68uF. Or, the good doctor simply picked two different types of caps from the box.
-
I tried this concept late last night, with a .047uf cap across the 100uf electrolytic in my current bench tweaker SE EL34. I also had a rotary switch with a bunch of values that I was able to switch through, but they were .01 and lower. I went back to the .047 because that seemed to sound the best of the bunch. I also tried a .1uf but veered back to the .047.
Disclaimer: It was after my band practice and I had some ear fatigue but I did seem to prefer the response with the cap in there. I've got a couple other builds laying around that I can experiment with and my initial report is that there is something here for us tweakers to tweak.
This specific SE build would get a little "ratty" at higher volumes and the cap seemed to help.
Nothing scientific here but it will prompt me to take the scope and signal gen. to it this weekend.
It is a simple-enough "try it and see" mod. Just be super-careful because we are talking about 400+ volts.
Thanks for bringing this up PharmRock. :thumbsup:
-
take the scope and signal gen. to it this weekend.
:thumbsup:
if you can't see it on a scope, you can't hear it with your ears.
sometimes though what a scope misses a spectrum analyzer finds, especially when looking at harmonics
the $2000 scope I have can be had for about $300 now, has an FFT mode which comes in handy for harmonics
-
I tried this concept late last night, with a .047uf cap across the 100uf electrolytic in my current bench tweaker SE EL34. I also had a rotary switch with a bunch of values that I was able to switch through, but they were .01 and lower. I went back to the .047 because that seemed to sound the best of the bunch. I also tried a .1uf but veered back to the .047.
I've read (KOC) where if you where to use all film caps for B+ filter caps in the PSU, in stead of lytics, you would hear a difference in the amps distortion, "something would be missing".
So, maybe that's why you liked the smaller value cap? Hi-fi guys use bigger than that, but then it's hi-fi.
I think it's jjasilli who puts the film bypass cap across the last B+ filter cap, to effect the 1st preamp gain stage.
-
While the sluthing hat is on, can anyone tell me why the green cap across V2 cathode resistor(820 ohm) is smaller than its brother across V1 2k7 cathode resistor, it was reported as both being 0.68uF.
I can’t see that part of the circuit, but One may be electrolytic and the other a film cap. I know Suhr does this on their SL68. Supposedly, a film bypass cap has a different tone but to get them in that value requires a huge (size) cap.
-
Hi,
Full gut shot.
John
-
my initial report is that there is something here for us tweakers to tweak.
------
Nothing scientific here but it will prompt me to take the scope and signal gen. to it this weekend.
Thanks for doing the preliminary test SG...looking forward to hearing some more details after you get a chance to look into it some more.
I have my Plexi50 on the bench right now. While I don't have a scope or a device such as you have to cycle through different cap values, I'll try to hook up a 0.047uf cap as shown on the EMS, as well as any other caps in that neighborhood that I have lying around and report any changes in the tone that I find. Hopefully I can get to it this weekend.
If I decide to put this on a switch (ala "touch" switch)...just to verify...the switch must go between cap and ground, correct? There should be no DC on the switch in this configuration.
-
If I decide to put this on a switch (ala "touch" switch)...just to verify...the switch must go between cap and ground, correct? There should be no DC on the switch in this configuration.
Correct.
Not that it MUST go there, but it's a wayyy better place for it.
I wouldn't call it so much of a TONE tweak, but more like a TEXTURE tweak (if that makes any sense)
It seems to tighten things up and take away some harshness. At least that was my first impression.
I'm looking forward to hearing your thoughts and getting more time to dive into it.
-
Browsing JTM45 schematics and happened upon this one from Mojotone. They have the 0.047uf cap in their schematic. See attached.
-
Several of the Fender Reissue amps use a .047.
https://el34world.com/charts/Schematics/files/Fender/Fender_59_bassman_manual.pdf
https://el34world.com/charts/Schematics/files/Fender/Fender_57_bandmaster_reissue.pdf
-
Alright...so today I wired in an ON-OFF-ON mini toggle in my 50 watt Plexi. Both caps were coming off the first 50+50uf filter stage, ran each cap to separate poles on the switch, with a ground wire in the center pole on the switch.
I tried a 0.047uf cap in one position and a 0.022 in the other. I had it pretty cranked and played some chords and single note stuff in different switch positions. Could hardly tell anything was happening. If anything, the 0.047uf cap maybe made things a little "smoother" if that means anything. I would let the chords ring out and change the toggle position from no cap to either the 0.022 or 0.047 and really didn't notice much. Honestly I think if I was blinded to the switch position I wouldn't notice any difference at all. Maybe my ears aren't trained for picking this stuff up, but I can't seem to justify having it on a switch. Worthwhile experiment, and learned something along the way.
-
You have to use a larger value cap. Those values are too small to hear a difference.
Try 0.47 to 1uf up to 10uF. They have to be film caps.
-
You have to use a larger value cap. Those values are too small to hear a difference.
Try 0.47 to 1uf up to 10uF. They have to be film caps.
I was scrounging the parts drawer but its a little bare at the moment...I didn't have anything in those ranges that were rated for 450+VDC. I'll keep the switch there for now, as I have some other tweaking to do anyways. In the meantime I'll get a variety of caps ordered and once they arrive I'll "plug and play" with a few.
-
You have to use a larger value cap. Those values are too small to hear a difference.
Try 0.47 to 1uf up to 10uF. They have to be film caps.
But isn't the cap in the Dr Z is a .047µF? Sure looks like it to me.
-
You have to use a larger value cap. Those values are too small to hear a difference.
Try 0.47 to 1uf up to 10uF. They have to be film caps.
But isn't the cap in the Dr Z is a .047µF? Sure looks like it to me.
Yes...as well as the Mojo JTM45 schematic and the Bassman reissue. Could be that the 0.047uf is more evident in these circuits vs the 1987 circuit?
-
I never though the .047s in those amps had anything to do with tone. Always figured it was to suppress high freq. noise spikes from the power line or rectifier diode switching. Kinda like the .047 "death cap".
-
But isn't the cap in the Dr Z is a .047µF? Sure looks like it to me.
True, but that's also in that amp build, might make a difference because of the PSU in that amp and demands on it. In a diffenert amp, Dr. might have used a different value for that cap, not a 1 size fit's all. Just guessing. :dontknow:
Not trying to compare apples to apples, he doesn't have the same amp. But trying to see if he can hear any difference by using a larger value. Experiment.
Might just be a feel factor thing, could be very subtle. Like upping/lowering, doubling/halving the filter caps uF value.
-
This is totally unrelated to the discussion at hand, but what the heck is the significance of the kinky bends in the wires? Is it just the "artists" little quirk to identify his work?
-
This is totally unrelated to the discussion at hand, but what the heck is the significance of the kinky bends in the wires? Is it just the "artists" little quirk to identify his work?
Good question. :w2: :dontknow: :laugh:
Some old Gibson amps they did that to the C and R leads, so they didn't run ||, less signal cross talk, parasitic, injection.
-
The kinks are for strain relief. I knew one tech that put little curly pig tails on ecaps for strain relief...
http://www.timeelect.com/jimsuper-33.jpg
-
In my earliest builds I employed a 100nF cap across the first filter—mostly because it was common to the Valve Jr at its many modified incarnations. Hey, we all start somewhere.
The last many builds I have abandoned this technique.
I was inspired to see if it made a difference on my latest build. It’s a single channel no tremolo amp inspired by the 6G3 but employs a 6K6GT PA. I’ve been fine tuning it trying to carve away some of the mud.
I clipped in a .1uF across the reservoir cap and it made a noticeable difference in clarity reducing some of the hairy fuzz. It made much less of a difference (virtually none) when place across the preamp’s filter cap.
The .047uF was not really effective enough to use it in this amp, but .1uF definitely made a difference.
-
Alright...so today I wired in an ON-OFF-ON mini toggle in my 50 watt Plexi. Both caps were coming off the first 50+50uf filter stage, ran each cap to separate poles on the switch, with a ground wire in the center pole on the switch.
I tried a 0.047uf cap in one position and a 0.022 in the other. I had it pretty cranked and played some chords and single note stuff in different switch positions. Could hardly tell anything was happening. If anything, the 0.047uf cap maybe made things a little "smoother" if that means anything. I would let the chords ring out and change the toggle position from no cap to either the 0.022 or 0.047 and really didn't notice much. Honestly I think if I was blinded to the switch position I wouldn't notice any difference at all. Maybe my ears aren't trained for picking this stuff up, but I can't seem to justify having it on a switch. Worthwhile experiment, and learned something along the way.
Nothing happens when you switch it while you are holding a chord because the initial transient surge is over.
Because it's not a TONE tweak it is difficult to pick up and identify, but from my experience there is a difference and I could see there being a value in either having the cap or not, for some guys. It makes sense to me that Dr. Z called it TOUCH.
I agree that it doesn't really justify being on a switch, but I've seen dumber switches.
I was only able to pickup a very slight blip on my scope and still didn't have enough time to give it it's due. I'm still not done with it.
-
Several of the Fender Reissue amps use a .047.
It may be significant that they’re on the rectifier side of a hot switching standby. And so by acting as a mini reservoir cap, may be taking some of the current surge away from the rectifier when standby is flipped to play mode.
So perhaps intended as a ‘be nice to rectifier’ cap, rather than a ‘fast transient response’ cap.
… Kinda like the .047 "death cap".
Bear in mind that the death cap’s purpose was to keep the chassis as close to ground potential as possible (ie the neutral wire), without passing too much current if it was switched to connect to the live wire. Not to suppress switching noise.
With polarised mains connectors, its on off on switch wouldn’t have been beneficial / necessary, with it switched to connect to the live wire, only a masochist would want to experience a few mA of mains current.
Of course these days, the live wire carries most of the hash noise, but it would be erroneous to conflate the death cap with a X type mains noise suppression cap.
-
but it would be erroneous to conflate the death cap with a X type mains noise suppression cap.
I wasn't conflating it with a class X cap. I was comparing to a class Y cap. Back in the two wire unpolarized day they worked very well to reduce some line noise.
-
but it would be erroneous to conflate the death cap with a X type mains noise suppression cap.
I wasn't conflating it with a class X cap. I was comparing to a class Y cap. Back in the two wire unpolarized day they worked very well to reduce some line noise.
Sorry, I’ve no experience of ungrounded, unpolarised mains, so was hypothesising :think1:
But in the absence of an actual ground connection, I don’t see how a class Y (ie live to ground) cap could be described as such? I suppose the chassis becomes a notional ‘ground’ in that scenario.
Or how flipping the ground switch so that the chassis was pulled up to mains live potential might reduce noise?
I thought that noise would automatically be lowest when the chassis was AC connected to ground (ie via the neutral earthing rod), rather than to AC live?
I can envisage that scenario would change with a grounded mains electrical installation and 3 wire power cable and grounded chassis. ie that the lowest noise setting of the ground switch would probably be as you describe, with the death cap connected between ground (the chassis) and the live wire.
-
Pardon my slowness but how is this not like a dull or smoothing cap on V1 plate? It’s essentially a very low capacitance non-lytic capacitor passing B+ to ground. Usually in Marshall styles it’s a 200-500pf silver mica. This is just a film placed off the first node instead of the last (preamp)??? :dontknow:
-
Pardon my slowness but how is this not like a dull or smoothing cap on V1 plate? It’s essentially a very low capacitance non-lytic capacitor passing B+ to ground. Usually in Marshall styles it’s a 200-500pf silver mica. This is just a film placed off the first node instead of the last (preamp)??? :dontknow:
A dull or smoothing cap on V1 does not pass B+ to ground. Neither does the cap in question.
B+ is DC.
-
Pardon my slowness but how is this not like a dull or smoothing cap on V1 plate? It’s essentially a very low capacitance non-lytic capacitor passing B+ to ground. Usually in Marshall styles it’s a 200-500pf silver mica. This is just a film placed off the first node instead of the last (preamp)??? :dontknow:
Have you got your head around HT nodes being at AC 0V?
That’s what the HT decoupling caps do, ie ensure a very low impedance between the HT node and 0V.
-
Pardon my slowness but how is this not like a dull or smoothing cap on V1 plate? It’s essentially a very low capacitance non-lytic capacitor passing B+ to ground. Usually in Marshall styles it’s a 200-500pf silver mica. This is just a film placed off the first node instead of the last (preamp)??? :dontknow:
Have you got your head around HT nodes being at AC 0V?
That’s what the HT decoupling caps do, ie ensure a very low impedance between the HT node and 0V.
So Jordan, by the time we are downstream at the filtered node for V1 we have 0V AC.
But, coming straight out of the rectifier we have some AC ripple voltage.
The cap you mention just shunts some highs from the AC signal to ground whereas this cap in question shunts AC ripple to ground.
That's what makes it thread worthy. A closer look at why the mighty Dr. Z thought this was worth a front panel switch.
-
So Jordan, by the time we are downstream at the filtered node for V1 we have 0V AC.
Cure ball. From Aiken amps web site;
"I also tend to put the preamp filter caps in the circuit where they are used ("local" bypassing). For example, the first preamp tube's filter cap will be physically placed in the area of the first preamp tube, with the ground and B+ connections made right to the bottom of the cathode resistor/cap and to the top of the preamp tube resistor. If I decide to instead group all the filter caps together, I'll be sure to also locally decouple each filter cap node with a smaller capacitor, typically a 0.1uF/400V) directly from the top B+ node of that stage to the ground node at the bottom of the cathode resistor. You'd be surprised how many brand new filter caps have very high reactances at frequencies within the range of a distorted guitar. Sometimes you can take a "bad" filter cap and bypass it with a 0.1uF cap and it will sound fine. It never hurts to have good high-frequency decoupling at all nodes."
-
Sounds like Mr. Aiken agrees with us.
Maybe we need a switch with a cap on each stage. :dontknow: :icon_biggrin:
-
To me, this stuff is getting closer to the fringe or maybe in the realms of "gilding the lily"?!?
Doesn't mean its not interesting and shouldn't be explored though!
There are other areas you could also explore plopping a film cap in (just to play devils advocate). What about the bypass of a cathode biased output section? I know some people have tried larger bypass caps (1000uf+) in that section to emulate fixed bias. Maybe a 1uf film in addition could add that extra "something".. or maybe just trying a big' ol fattie film cap in there might be interesting. :think1:
-
maybe start with a SS PS, tweak values for tubes, then remove one cap at a time n evaluate :icon_biggrin:
-
Here's the link;
https://www.aikenamps.com/index.php/grounding
-
So Jordan, by the time we are downstream at the filtered node for V1 we have 0V AC.
But, coming straight out of the rectifier we have some AC ripple voltage.
The cap you mention just shunts some highs from the AC signal to ground whereas this cap in question shunts AC ripple to ground.
That's what makes it thread worthy. A closer look at why the mighty Dr. Z thought this was worth a front panel switch.
Sounds like Mr. Aiken agrees with us.
No, I don't so. I think it's another thing. Curve ball. It's only the "very high reactances at frequencies within the range of a distorted guitar". Heavy distortion generates a lot of very high frequencies that some lytic caps can't decouple from the B+.
I don't think he's talking about ripple on the 1st stage, there's not enough of it there.
-
Sounds like Mr. Aiken agrees with us.
Maybe we need a switch with a cap on each stage. :dontknow: :icon_biggrin:
I think you may have missed the highly sophisticated sarcasm in that post?
I don't think this is a mystery anymore.
Let's give it a name:
Output Stage Transient Snubber
And a new abbreviation: OSTS
sluckey, please add that to the Official Amp Acronym List or OAAL as I like to call it
-
I think you may have missed the highly sophisticated sarcasm in that post?
Well, that's because I would have to be highly sophisticated. :l2:
Let's give it a name:Output Stage Transient Snubber
But it doesn't snub the transient, it alows them by responding faster that the lytic.
And a new abbreviation: OSTS
sluckey, please add that to the Official Amp Acronym List or OAAL as I like to call it.
NO!!!!!! NO MORE Acronyms!!!!!!! :BangHead: :cussing:
I can't remember 3/4's of them now! :BangHead: :cussing:
There's a reason we use letters to make words. Everything is backwards these days!!!!!
-
Ok, lets just call it the Dr. Z Touch Switch. :icon_biggrin:
Or DZTS?
-
The Ditzs switch? :undecided: :l2:
-
Let's give it a name:Output Stage Transient Snubber
But it doesn't snub the transient, it alows them by responding faster that the lytic.
One of us is off here and it makes me nervous that pdf64 liked your post. I looked up the definition of snubber in case I was using it incorrectly.
I used the term "Snubber" because:
"Snubbers are used in power circuits to suppress harmful voltage transient spikes"
-
Edit; I'm talking about the ac signal component, its transients, not the dcv. Read the Aiken quote again.
Let's give it a name:Output Stage Transient Snubber
But it doesn't snub the transient, it alows them by responding faster that the lytic.
One of us is off here and it makes me nervous that pdf64 liked your post. I looked up the definition of snubber in case I was using it incorrectly.
I used the term "Snubber" because:
"Snubbers are used in power circuits to suppress harmful voltage transient spikes"
That's what I'm saying, I think you have it backwards.
Lytic caps have flaws in their ability to filter the very high and low ends of the frequency spectrem. After a certain point, the higher and lower the frequency goes the less able they are to filter those frequencies. And high distortion creates lots of hi end harmonics.
The film cap can respond much faster and in the high frequency range that lytics can't. So the initial pick attack, spike in the ac signal, is not rolled off, it is not snubbed.
-
I’m sceptical that the cap could have anything more than a negligible effect on the amp’s transient response.
-
Hi-fi guys have been doing it for years. (Yes they do some wacky things.)
Same type of thing as using all film caps for the B+ filtering. KOC and some of the guys that hung out at his old forum, built amps with all film caps and talked about the difference in sound and response/feel. And like I said Carr amps does the same with a few models. His amps are highly respected.
-
Indeed but no matter how venerable the source, it still only seems to amount to so much idle hypotheses. And highly respected builders don’t have to be EE superbrains.
Transient response is something that can be characterised, analysed and differences be made quantifiable. Isn’t that what the impulse response testing of the different speakers is all about?
The resources to do that are beyond the scope of most DIYer’s test bench, but surely, if so minded, those people could put some evidence behind their hypothesis?
-
Hi-fi guys have been doing it for years.
the A/B switching in sand is different than glass, therefore a different "take" (see schematic reply #51) on making DC stay DC
one of the "ball buster" cal's I had to accomplish, make the DCV go from 0 to 320V in 60mS. measure n meet spec at all 5 TC's (time constants) on the way UP. once that was done, make 320VDC stay 320 VDC +/- .5v for ANOTHER 5 TC's (0-200(IIRC) seconds)
any time we switch Volts or current on/off fast, especially into inductors, (coils), transients can n do happen, in music that might produce a happy "sound". In constant source DC, they are never welcome. N, ya PDF64, the math is crazy and doesn't help AT ALL for calibrations
-
None of this matters if you can't hear it of feel it when you play. Have to experiment to see what you hear. Not a big deal.
The evidence is what you hear. They wouldn't build amps with poly film power supply caps if they and their customers didn't hear anything different.
If an audio amp is basically a modulated power supply, then the PSU is the foundation of the amp. So the PSU would have a big influence on an amps overall sound and performance.
Speed players complain about being able to out run certain amps, PSU.
Still, with bypass film caps in the PSU, it's probably pretty subtle, not a massive change, as it's only a small change to the whole PSU. Changing to full film caps, not just a small bypass film cap, would probably be more noticeable.
-
...
Speed players complain about being able to out run certain amps, PSU.
...
That seems like another example of hypothesis that I query the credibility of.
-
That's been long expressed by speed players. They change amps because of it.
It's the sag and compression, they want more articulation. So they go to amps with a stiffer PSU, SS rectifier, larger filter caps, even going to 6550's. Larger wattage speakers, less compression, more articulation. Larger PT B+ reserve, mAs, larger OT's. Any/all of those make can help make for a stiffer amp.
They don't need the sag/compression, they play fast note's not long sustaining bent string note's that bloom, like blues/rock players.
I had a Mesa MkII C, never liked it, way too stiff for me, too articulate, I could hear the smallest little pick attack thing, extremely unforgiving. The least little subtle pick mistake, you could hear it. Now some guys would love that. And I played a BF SR for years, main amp, not exactly a weak PSU in those those amps. But certainly more grace than that Mesa.
-
You'd be surprised how many brand new filter caps have very high reactances at frequencies within the range of a distorted guitar. Sometimes you can take a "bad" filter cap and bypass it with a 0.1uF cap and it will sound fine. It never hurts to have good high-frequency decoupling at all nodes.
This is a super interesting thread. However could someone please explain to me how the reactance of a cap at the B+ node could interact with the frequencies of the signal? I don't understand that.
-
…how the reactance of a cap at the B+ node could interact with the frequencies of the signal?
Standard circuit analysis assumes that power supplies have zero impedance to 0V circuit common.
High value caps (usually ecaps) are used to achieve something hopefully close to that in real world. But if the impedance is not so low as to be negligible, then as 2 or more stages might share the same power supply node, and the impedance of that node becomes significant, then unexpected interactions between those circuit stages will occur. Because signal paths occur where none should be.
It’s a similar scenario to the circuit board losing its insulative property.
-
Electrolytic caps are great for filtering out low frequency 120Hz and even audio signals from the B+ line. But their response time is too slow to effectively remove high frequency stuff such as noise spikes that come in on the power line or rectifier diode switching spikes, or square waves, etc. Small film caps have a much faster response time to high freq stuff so they are much more effective at removing high freq noise spikes, etc from the B+ line.
-
…
It's the sag and compression, they want more articulation. So they go to amps with a stiffer PSU, SS rectifier, larger filter caps, even going to 6550's…
Thanks, whatever makes people happy I suppose!
A vintage type amp still works fine at full load though, can put out continuous max power until they run out of notes or whatever.
The HT supply doesn’t differentiate between a continuous single note and an continuous stream consisting of different notes; full load is still full load, I don’t see how the HT cap type would make a difference?
-
^^^^
agree, I think one "magic genie" is current demand over a short time cycle, a PT that just barely meets the load requirement verses one that is 2X the load required would go further on sag/response than a cap time constant. Transients are better dealt with with fast response caps as Sluckey pointed out
-
A vintage type amp still works fine at full load though, can put out continuous max power until they run out of notes or whatever.
The HT supply doesn’t differentiate between a continuous single note and an continuous stream consisting of different notes; full load is still full load, I don’t see how the HT cap type would make a difference?
That's not what I said, it's not about full power. It's about sag, compression and not rounding off the very front end, the initial attack, of the notes envelope. I also said SS rectifier. You can still have full power and be well into the amps PSU sagging, the OT compressing the far ends of the frequency spectrum more and more and the speaker into compression too. In fact you get more power as the sine wave squares off. As this happens the very 1st pick attack gets more and more lost, diminished, yet you can still have full power output.
It's a matter of feel. You have to get behind the wheel and play through the amp. Some love 'grezzy' amps, weak PSU, squishy/saggy amp. If you can have that end of the spectrum of an amps feel, you can have the other end too. Amps are tools, and as such can be under-powered or over-powered for the job.
A friend of mine and I were A/Bing speakers 1 time. When he was playing I could hear some difference in sound/tone, but nothing drastic. But when I got behind the wheel, there was a huge difference in how the speakers responded to pick attack.
Another time, had a BF Dual Showman, had a gig, set up, turned on the amp, something was wrong. Sounded odd. There was a guy who came to see us and sit in. He was playing 2nd guitar for Buddy Guy and Jr. Wells, had played for them for several years at that point. He got up and played through my Showman, after I asked him if he thought the amp sounded odd, he thought it was fine. :dontknow: And he was a Fender BF guy. :laugh: I took the amp into the shop, before I worked on amps. One of the 6L6GC's died. I heard it, he didn't, my bass player didn't. The bass player played with me for several years, so he knew that amps sound.
Depends on how you play, what style you play and how developed your style and ears are.
I had a Mesa MkII C, never liked it, way too stiff for me, too articulate, I could hear the smallest little pick attack thing, extremely unforgiving. The least little subtle pick mistake, you could hear it. Now some guys would love that. And I played a BF SR for years, main amp, not exactly a weak PSU in those those amps. But certainly more grace than that Mesa.
How do you explain this? This was way before I had read anything about PSU's, sag, etc.
Do you play? It's in the hands, ear, not paper that you get to know these things.
-
It's in the hands, ear, not paper that you get to know these things.
Along these same lines, my shipment of caps arrived earlier this week (Thanks Doug for quick shipping) and I plan on moving up to a 0.1uf cap on the switch to see if I notice anything. Maybe try a 0.047 to 0.1uf in parallel. I'll report back what I find.
Thanks to all for the comments and replies on this thread. I've been doing my best to keep up with the more technical aspects of the discussion and have learned some things along the way.
-
Since it's an experiment, and you have the film bypass cap on a switch, I'd try it at different volumes, even if you don't normally play at those settings. Leave the tone settings the same on the 1st test. Looking to see if PSU sag plays into this in any way, it might. :dontknow:
I'd try starting at low volumes going to higher setting, so your ears don't fatigue.
Maybe start at 3 flip to bypass, then 4 flip to bypass and so on all the way up to 8 or more. I'd also try it latter again, ear fatigue recovery, with the treble set higher. It might revel it self clearer in the higher frequencies.
And I'd also be listening to hear if the very front end of the note, initial pick attack, changes, is clearer, crisper, more percussive, responds differently to your pick attack, etc.
I was thinking and wondering if maybe, that at least part of what the bypass cap does, if it does anything, is removes enough/all the high frequency line noise that was masking the initial pick attack? :dontknow:
Take your time, it may be subtle. It may be more feel, touch response, than heard, or it maybe some of both. :dontknow:
-
This morning I soldered in a 0.1uF cap on the switch. Amp was set up for high gain, but this amp cleans up nicely with the guitar volume control. I used a Washburn superstrat with a JB and Jazz, and also a partscaster with Lollar Blondes. The amp went into a reactive load with headphones attached so I could hear the details better. To minimize any bias, I had my wife switch the cap in an out while I played, with me not knowing whether it was switched in or out of the circuit. Towards the end she let me know whether the cap was switched in or out.as she operated the switch.
To my ears, there is a subtle but noticeable difference with the 0.1uF cap switched in. The pick attack was more clear and the top end less ratty/fizzy, especially on the higher notes. Not sure if I would call it smoother, but as SG said previously, not as harsh. It was more evident at higher gain settings. But even with low gain there was a noticeable difference.
Currently the switch is located on the side of the chassis. I think I will leave it there for a while and if it seems like I am using it a lot, I may relocate it to the front panel, or maybe convert one of the tone controls to a push/pull.
One more experiment is to switch the 0.1uF and 0.047uF in parallel, just for curiosity's sake. I'll try that later today and report back.
-
That test is fair in 1 way and not another. I believe you are hearing a difference, but can you hear the same difference through the speaker as you are hearing with head phones? The speaker might mask or filter what your hearing? Have to try the same test through the speaker you have.
Did you try the test you did with the .047 cap? I wonder if you could still hear it with the smaller value cap?
That caps value is way to small to affect the PSU's ripple, and the ripples frequency is to low, so it has to be affecting some other component of the sine wave or ridding on the sine wave.
Edit: I mean the total magnitude of the ripple.
-
Let's give it a name:
Output Stage Transient Snubber
And a new abbreviation: OSTS
sluckey, please add that to the Official Amp Acronym List or OAAL as I like to call it
Official Forum Amp Acronym List would yield a somewhat better acronym.
-
That test is fair in 1 way and not another. I believe you are hearing a difference, but can you hear the same difference through the speaker as you are hearing with head phones? The speaker might mask or filter what your hearing? Have to try the same test through the speaker you have.
Did you try the test you did with the .047 cap? I wonder if you could still hear it with the smaller value cap?
That caps value is way to small to affect the PSU's ripple, and the ripples frequency is to low, so it has to be affecting some other component of the sine wave or ridding on the sine wave.
Edit: I mean the total magnitude of the ripple.
So its been almost a month, and I've been playing the amp pretty much cranked through a reactive load/attenuator, listening either through headphones or a more neighbor-friendly volume through a 2x12 closed back cab (amp is still cranked, but attenuated). I changed the switch to a On-Off-On, so I could compere either the 0.1uf or the 0.047uf to no cap at all. I'd play, mess around with the switch, play some more. Then come back the next day not remembering what position I left the switch in and playing some more (switch is mounted on the side of the chassis...can't easily see it).
I am just not hearing a whole lot of difference overall especially through the speakers. It is very subtle at best.
After doing a little more digging, similar to what Sluckey found about the Bassman reissue that has a 0.047uf cap, I discovered that the JTM45 schematic dated 2001 (reissue?...see attached) has a 0.05 right off the rectifier to ground. I've come across this before with the Mojotone JTM45 kit schematic, and it is shown on the attached PDF as well as a schematic I found on Marstran website.
-
After doing a little more digging, similar to what Sluckey found about the Bassman reissue that has a 0.047uf cap, I discovered that the JTM45 schematic dated 2001 (reissue?...see attached) has a 0.05 right off the rectifier to ground. I've come across this before with the Mojotone JTM45 kit schematic, and it is shown on the attached PDF as well as a schematic I found on Marstran website.
Yes, Marshall was doing it years ago, Kevin O'Connor has it in his books from the 90's. But it was for B+ line noise like Sluckey said.
-
Didn’t Marshall just copy it off the 5F6A?
My guess is that it was intended to take some stress off the rectifier from the hot switching standby.
-
Didn’t Marshall just copy it off the 5F6A?
My guess is that it was intended to take some stress off the rectifier from the hot switching standby.
Fender shows a .05uF cap before the standby.
How is that small a value cap going to take any stress off the standby switch?
-
Fender shows a .05uF cap before the standby.
So does the JTM45, it looks the same to me :icon_biggrin:
How is that small a value cap going to take any stress off the standby switch?
I was thinking the rectifier rather than the switch. ie the little cap will charge up to the peak unloaded HT winding voltage in standby. Then when switch is closed, the little cap will dump its charge into the main reservoir, so the initial peak rectifier anode current won’t be quite so high.
Ok, it’s tenuous, but it’s my hypothesis and I’m sticking with it :think1:
-
I was thinking the rectifier rather than the switch. ie the little cap will charge up to the peak unloaded HT winding voltage in standby. Then when switch is closed, the little cap will dump its charge into the main reservoir, so the initial peak rectifier anode current won’t be quite so high.
Ok, it’s tenuous, but it’s my hypothesis and I’m sticking with it :think1:
Ok, that's your story and your sticking to it. :icon_biggrin:
Oh, yes the diodes. I read that wrong.