I say weird because I have my doubts as to whether it will work.
Once I've finished messing with my effed up computer I'm having trouble with. I'll Breadboard it.
Meantime pedals are a poor second to getting my IT back up and working.
Here's a schematic
http://www.zimagez.com/zimage/schema79.php
Edit
Does anyone see a good reason why I shouldn't replace the 2x 2.2M bias resistors with a noiseless bias scheme and a 1M from Vbias to IC1a +input pin?
Anyone built it?
See any reason it will work or not?
Cheers fellas
>Does anyone see a good reason why I shouldn't replace the 2x 2.2M bias resistors with a noiseless bias scheme and a 1M from Vbias to IC1a +input pin?
No - that should work fine.
>See any reason it will work or not?
Haven't built it, but it looks like it will work, although I would use a single-supply op-amp rather than a TL072. I would not expect stellar performance.
Cheers Sam.
Breadboard project it is then.
Any improvements that might make it Stella?
subbed, out of curiosity...
just curious as to the results...whether it just cuts off instantly , or has fair adjustability..
my breadboards are all full at the mo, otherwise i'd try it...
there are some decent Q&D gate designs with OTAs à la R.A. Penfold, just sayin'
Thanks Gary that maplins noise gates on my list Too. I bought some chips before Xmas.
Just canvassing for opinions before I bread it. I'll compare to the DoD Fx30 too.
There is a built-in issue with using MOSFETs for a variable resistor. They're VERY good at this, but the inherent body diode acts like a silicon diode from source to drain. So for any signal up around 0.4V or bigger across the MOSFET, you get the beginnings of conduction in the body diode.
Boss has used a similar circuit, but rearranged to use a JFET instead of a MOSFET for the shunting element and take care of the opposite DC offset of JFETs.
In the schemo, IC1B biased at ground will have the potential to cause ugliness with the TL072 in particular, as this is outside the input common mode range of the chip. It might work OK, might not, or some makers' TO072s may work, others not. If you have issues, put in a dual opamp that has an input range that includes V-.
I would be very tempted to take IC1B's signal input from the output of IC1A through a cap, as this would lessen any input loading.
>Any improvements that might make it Stella[r]?
RG largely beat me to it regarding tweaks/issues.
Another advantage of taking the comparator after the input buffer is that you could give the buffer some gain - you will have to see how well the triggering works, but it might help to have a hotter (or colder) signal.
If it takes too long to recover after triggering - which it might, it should be easy enough to make the recovery adjustable.
It is also trivial to add a 'triggered' LED if you want to see when the gate is opening.
There's something wrong around the output isn't there? Q1 will shunt the output to ground through R10 but it looks like R10 is 22meg so it's not going to do anything useful, or am I looking at it wrong. Also what are R12 and C9 supposed to be doing, I wonder if it's drawn wrong and the pot is supposed to be a volume control.
On the plus side the Russian girl who wants to chat to me looks nice :)
>There's something wrong around the output isn't there?
It's a little weird, but looks like it will work after a fashion.
I read R10 to be 2M2. In any case, when the gate is closed, the output is LP filtered with R12 pot (0-470k) vs C9 (100n). This will get weird with R12 nearly CCW (low resistance), but otherwise is shunting the audio band to ground.
When the gate opens and the MOSFET conducts, the FET drain || R12 resistance becomes negligible and the corner frequency shifts way above the audio band and you pass signal.
Edit - Rich, you should try making C9 somewhat bigger if you find that it is not gating bass very well - try 1-10u.
Cheers, yeah that makes sense now.
Thanks guys this is at the back of a long queue. I'll get around to it
Just bread boarded it as per schematic and i cant get it to work yet. Its late and time to hit the hay so i'll have a gander in the morning revisit it and try some of the suggestions youve all made
Night all
finally back at the BB
i have to take out c9 for ic1a to even make noise.
now for ic1b
its gating
things that made it better
taking the comparators input from the output of Ic1a through a 100nf cap I may need to increase or decrease this
changing c9 for a 1uf
its gating but the cut off is too fast most of the time other times its on off on off off
getting there
thanks everyone
for C9 ive tried 10uf, 1uf, 470nf, 100nf
1uf seems to work the best
pot1 set to 16.1>16.3k
the mosfet shuts off quick depending on where the pot 2 is set Im calling this gate.
i guess pot 1 sets the dc voltage for the IC1b its a very narrow margin with setting this up and with the pot 1 set up at above K ohms setting its dc is 4.3>4.7volts
Is there a way to slow down the mosfet on off time?
maybe changing C7?
maybe a further series diode with D3?
Maybe a differnat D3? Germ, led, another?
At some settings the mosfet is not sure if its on or off. It makes a distorted kinda diode conducting noise.
All trials with no other effects just my MIM Strat.
Lord knows if it will behave with another pedal before it.
As it is I'm thinking last on the pedal board because of the mosfets quick shut off.
BTW input for IC1b is from the output of IC1a through a 100nf series cap.
IIRC RG mentioned something about Boss usng a Jfet instead. Do i have to add any further componants or just replace the mos with a J Fet?
Thanks Guys
If this works out it may be the cheapest low parts working guitar clamp I've ever seen and I'm using a TL072.
No Pinching my Clamp name either. :)
Boss has used a similar circuit, but rearranged to use a JFET instead of a MOSFET for the shunting element and take care of the opposite DC offset of JFETs.
So can I just use a jfet without any further parts.
Thanks
It might work without modification with a P-channel JFET.
Otherwise, you will have to invert the comparator's behaviour.
Cheers I'll try those.
Heres where I'm at.
it gates but the gate comes on really fast. too fast for use.
primeraly this is the issue.
I tried inverting the comparator and using a jfet tried 2n5458, j201, J113 similar results to before
I tried a pnp mosfet BS250 similar
Any ideas on how to control the note decay to switching a bit slower?
other things i tried for kick
a 1n914 from drain to gate
a 1n914 from source to gate
a 1n914 from D3 cathode to gnd both ways round
the comparator input is taken from the input buffers output through a 100nf cap.
would changing the cap value here help?
Thanks for your time
Frustrated country boy
Rich
Edit
further suggestions and ammendments are welcome
Schmitt trigger will this be useful. Switching chips ?
The time it takes for it to gate once the input falls below the threshold is set by C7 and pot2, try making C7 bigger.
>it gates but the gate comes on really fast. too fast for use.
I'm guessing it is supposed to be a noise gate. If so, then it is supposed to come on fast.
Is the issue more that it is too twitchy, such that it keeps cutting out/back in again?
Will do Slacker
Thanks
Quote from: samhay on January 28, 2016, 04:14:37 PM
>it gates but the gate comes on really fast. too fast for use.
I'm guessing it is supposed to be a noise gate. If so, then it is supposed to come on fast.
Is the issue more that it is too twitchy, such that it keeps cutting out/back in again?
Sorry I should have been more specific.
If pot2 is at full rotation the gate is on. I hear zilch.
If pot 2 is at 4pm it's glitchy. On off on etc
I can play with the other pots but that only somewhat shifts the position in which it gates ie 3 to 4pm maybe.
I have built the dod fx 30 (it's a modded one) the schematic's over at the other place.
Tuneable enough for me threshold and sensitivity. But this beast is not tunable.
This is a valuable lesson to me. I'm just getting my head around how the comparator works after a quick browse online earlier today.
I still have a lot to digest in this hobby.
Cheers
The comparator has HIGH gain.
The MOSFET may have HIGH gain.
Between the two, the on/off action is too abrupt.
Maybe lower the comparator's gain (actually work it as a linear gain stage, not a comparator).
Is this your own invention or plagiarized?
Thanks for the response PPR.
Not plagiarised it came from here
http://techniguitare.com/forum/realisation-montage/tranquility-noise-gate-malheur-composants-t8976-90.html
It was apparently an updated version of the Hot Amp Noise Gate that was floating around on the net somewhere.
I can't find it now I want to of course.
I'll look into lowering the (comparators) gain and mess with that.
All help great fully received.
Right having looked over the schematic (post1) I can't figure out how to lower the gain of the comparator.
I'll refer to it as comparator for the duration of the thread so we all know to which bit we refer even if it is n fact a linear gain stage.
Thanks for that PPR.
Since normally the gain of an op amp (in distortions) is set by the input resistor versus the feedback resistor.
In this arrangement there's no feedback resistor. I'm gonna google differential op amps mean time and see if I can figure it out.
I only have one other fet to try and I think it's a power mosfet (lrf520) so I doubt it's gain will be lower than a 2n7000 or BS170.
What y'all think.
power MOSFETs are for high current/voltage, not extra gain, no?
Could be a candidate then thanks Garcho. I'll go over the datasheet thanks.
Since PPR said it was a linear etc. I looked up comparators and a lot of the ones I've seen have dual
Bjts on the output.
Like the single power supply comparator equivalant circuit here..
http://home.cogeco.ca/~rpaisley4/Comparators.html
What's that all about?
I looked up differential op amp circuits and now understand that a comparator and differential are differant beasts. One more bit of knowledge helps.
https://bytebucket.org/intelligentagent/replicape/raw/6a578f24a5f4e12c159f60fb76229215ae0ee9aa/Doc/Literature/sloa067-1.pdf (https://bytebucket.org/intelligentagent/replicape/raw/6a578f24a5f4e12c159f60fb76229215ae0ee9aa/Doc/Literature/sloa067-1.pdf)
not sure if that will help, but while it's on your mind...
Nice read thanks
That explains Pauls comment.
You could try replacing the comparator with the gain stage from something like the nurse quacky http://www.home-wrecker.com/nurse-quacky.html (http://www.home-wrecker.com/nurse-quacky.html) I think this is basically what Paul was getting at with "actually work it as a linear gain stage".
Whip out everything from the bottom left of the schematic up to and including the diode on the output of the comparator Splice in everything from the sensitivity pot on the nurse quacky up to and including the LED on the output of the lower opamp, the sensitivity pot become the threshold control.
Got it thanks Slacker.
Thank you all that will give me something to play with tomorrow on the breadboard.
Thank you for your patience.
> power MOSFETs are for high current/voltage, not extra gain, no?
MOSFETs cover the range from small (30V 50mA) to huge (1,000V 100A).
They are handy for BIG work because they have "almost" as much gain (low drive requirements) as BJTs, but are much more robust (no second-breakdown).
They are handy for small work because the gain is more linear than a BJT, and they can conduct both-ways which a BJT isn't as good as.
_I_ think that plan is flawed. When any signal peak exceeds the comparator threshold, that MOSFET will flip all the way to ON. That's too abrupt for music. We want a small but non-zero delay before opening-up, to ignore minor noise-crap. And a lingering after sound "stops", of course, so tails are not chopped. It wants an envelope follower, not a comparator. I'd advise studying all the noise gates you can find. (Sorry, I don't have a good example; they all have flaws.)
Quote from: slacker on January 29, 2016, 12:08:38 PM
You could try replacing the comparator with the gain stage from something like the nurse quacky http://www.home-wrecker.com/nurse-quacky.html (http://www.home-wrecker.com/nurse-quacky.html) I think this is basically what Paul was getting at with "actually work it as a linear gain stage".
Whip out everything from the bottom left of the schematic up to and including the diode on the output of the comparator Splice in everything from the sensitivity pot on the nurse quacky up to and including the LED on the output of the lower opamp, the sensitivity pot become the threshold control.
this was the nearest ive got to getting it to playball with the above.
trouble was once the mosfet went off it wouldnt reopen when notes are played so i figured it must be the transistor holding its charge/discharge. something i read led me to put a shunt resistor to ground from the gate. that didnt work either. no matter how i twiddled the knobies i couldnt make it work properly.
i tried a Duck Arse envelope detector intended to light an led in the last op amp stage and did the mosfet switch as per schematic page one. No dice
i tried a fet as a series switch driven by the same DA env detector that was a fail too.
mmmhhh
Six hours on the breadboard and im no nearer to succes.
I'm jacking it in calling it a night maybe tommorow ill succeed.
Update/rant over
Update 2
I set it back up on the Breadboard as per schematic but inadvertent left out the 470k on pot one.
Setting the pot 1 so the voltage on the wiper is 0.036v seem to work better. It still cuts off stark but will allow low notes to decay before gating, however it's a bit tricky with the higher E string. Suspect because it's a lower strength signal. I'll do some more tweeking with the pot 1 and see if I can't overcome it.
It's looking like I'm gonna have to use a 25 turn trimmer for pot 1.
On I go.
Question for the knowledgable.
would a dedicated comparator chip such as LM393, LM311 be more useful than the op amp as a comparator in this circuit
thanks
https://www.fairchildsemi.com/datasheets/LM/LM393.pdf
http://www.ti.com/lit/ds/symlink/lm111-n.pdf
Oh and I'm getting a bit of switching noise
a ploppy compressed note sound on playing and sometimes a doddery on off on off when the gate closes (on)
suggestions for the elimination of this would not go astray.
ive changed the charging cap to a 4u7
the comparators bias network to +9v >100k > 100k pot > inverting input. seems like i have a more useful range on turning the pot.
Im using a 2k7 shunt resistor on the mosfet gate.
changed the value of pot2 to 1M
Changed R4 to 1M
and as before Im using a 1M resistor from Vbias to the input buffer NI input
the comparators NI input is fed from the input too as opposed to the earlier suggested buffer output to comparator input
Did I say I'm using an lm358 op amp?
I'll post up a drawing later if i get the chance
My better half is making me a Honey Do list as I type....... Happy days.........
Still hacking.
Join in if you like.
what does your circuit diagram look like at the moment? it been three pages, after all.
gate version 103
op amp lm358
super bright blue 5mm leds
im getting a distorted output like a gate after a distortion on most setting but it gates quite well
thanks
(http://i.imgur.com/WvXVUx6.png)
what's pot 4 doing? it looks like it's making a low pass filter with C6. and what's the point of R1?
Thanks Garcho. That was just the impetus I needed and the information to push this project on further.
So Now I have a low parts count gate that works on the breadboard.
Thanks everyone.
(http://i.imgur.com/HC28XKz.png)
check it out.
cool, glad you persevered with this...
hows the adjustability?...threshold?...demo?
The 500k sens has been renamed pad.
The 1M has been renamed Decay.
Flat out on the decay gates so as no signal gets through except a distorted sound. It's not without its problems I admit.
It's best to set the pad and decay to nothing to start off. This I assume charges up the charging cap C5.
It was a stroke of luck Garcho asking what the pot did it made me rethink, I remembered reading that a resistor across the charging cap changed the time it takes to charge /discharge.
I tried it with 2n7000 that works too but the on off of the mosfet switch is abrupt and falters.
You can change the value of C5 I found a 33uf the most useful because it allowed the note to decay more naturally before the gating.
I've not yet done trials with buckers. And only a small trial with 1 distortion pedal. There's more work to do to refine it I'm sure.
But it's low parts count. It gates.
I'm amazed I got it going with only a week or twos work.
I has a slight swell on some settings too.
I've just completed my perfboard s when I can I'll do a demo.
Thanks for the interest.
I'm happy with this now. Just trialed it at a decent volume and it's stella.
Theres no gate control, just a pad to pad down incoming signals and a decay pot for setting up your decay time.
It will gate once the note or chord dies away. How quick depends on the decay and pad settings.
If the decay is set to max you wont hear anything because the gate is always closed.I plan to remedy this with a series resistor on the decay pot somewhere in the region of 50-100ohms i will do some trials.
There you have it a simple to build noise gate with low parts.
Please leave a report here if you breadboard it or build it.
I have a verified perf board layout if anyone wants it.
Rich
further experimentation shows that a resistor in series between gnd and the decay pot stops the horrible fizzy sound i was getting when the decay pot is set to minimum decay.
i tried from 50 ohms up and i've settled on a 2k7. I'm sure other values will work too.
A further observation is that it will work with 1n914 instead of the leds. Itried some bat41 schottkys and they even worked. although not as smooth as the leds so im sticking with those. Leds in use 5mm blue (not the super brights) the ordinary ones.
Yep another senior moment. I forget the proper term for normal ones.
I'm thinking I'll rename this Down Time.
As that's more how it feels. The gate shuts when play stops.
I'll try to do a demo tomorrow guys.
cool name DT. like it.
look forward to the clips man.. 8)
Quickly demo
https://youtu.be/IQYkiPgUNHI
god, i hate this new net bullshit... graphics of course gone, and youtube says the vid doesn't exist.
glad ya got it workin rich, if ya would be willing to share what ya finally came up with, would love to see it.
i thought it was hotamp's noisegate. i remember whipping up a vero a couple years ago, it didn't work so i circle filed it.
i haven't revisited techniguitarre in a long time, maybe i should go back and have a looksee.
glad ya persevered bro!
I'll pm you Jimi.
When I get home from vacation .
thanks rich!!
enjoy the time away!
;)
No Down Time while on vacation?