DIYstompboxes.com
DIY Stompboxes => Building your own stompbox => Topic started by: tddy934 on January 28, 2010, 12:00:04 AM
-
I want to build a simple noise gate but can't find a schematic anywhere, the only ones I can find are super complicated.
I'd appreciate schematics, layouts, guides, ect.
-
I have 8 right off hand and tons more in my archives. It's been years since i posted an image and don't know how. to be honest it's 2:00am and i don't feel like using the "search" option to learn how to post images right now. if someone wants to give me a quick tip on how to post schems great. if not i'll e-mail them to you. I have:
DOD FX30 Noise Gate (looks easy)
Boss Slow Gear SG-1
Korg Noise Gate
MXR Noise Gate (v2)
Distortion Box (distortion + noise gate)
Gallien Krueger Noise Master 2000CPL
MXR Noise Gate (v1)
Noise Gate NF-1
Also, you may want to read some of Tom Scholz's patents. Some killer info.
-
go to imageshack.org, upload the picture(assuming its a picture), when it gives you a direct link, copy that link, and come back to this thread
click reply, and click the little icon in the second row down, second from left and paste the url into there
-
Tonepad has the MXR noise gate with all the stuff you need to make inc build reports.
-
I have 8 right off hand and tons more in my archives. It's been years since i posted an image and don't know how. to be honest it's 2:00am and i don't feel like using the "search" option to learn how to post images right now. if someone wants to give me a quick tip on how to post schems great. if not i'll e-mail them to you. I have:
DOD FX30 Noise Gate (looks easy)
Boss Slow Gear SG-1
Korg Noise Gate
MXR Noise Gate (v2)
Distortion Box (distortion + noise gate)
Gallien Krueger Noise Master 2000CPL
MXR Noise Gate (v1)
Noise Gate NF-1
Also, you may want to read some of Tom Scholz's patents. Some killer info.
whats the diff between MXR v1 and v2?
oh yeah for images, click the "insert image" button above the smilies (2nd from left), the open a new tab and get the IMG address from
photobucket or wherever you keep your images (i couldn't figure out how to upload from my pc), then go back to your message and stick that address
in between these >> (http://) (they will come up after you click 'insert image'. hope that helps! :)
-
go to imageshack.org, upload the picture(assuming its a picture), when it gives you a direct link, copy that link, and come back to this thread
click reply, and click the little icon in the second row down, second from left and paste the url into there
Thank you
-
here is one: http://hammer.ampage.org/files/LPM-23-NoiseGate.PDF
here is another: http://hammer.ampage.org/files/GainesNoiseGate.PDF
-
(http://yfrog.com/j3mxrnoisegatev1g)
http://yfrog.com/j3mxrnoisegatev1g
MXR v1
http://yfrog.com/14mxrnoisegatev2j
MXR V2
-
(http://img40.imageshack.us/img40/2674/mxrnoisegatev2.jpg)
MXR V2
(http://img687.imageshack.us/img687/4181/mxrnoisegatev1.gif)
MXR V1
-
here is one: http://hammer.ampage.org/files/LPM-23-NoiseGate.PDF
here is another: http://hammer.ampage.org/files/GainesNoiseGate.PDF
the LPM-23 seems like a fun pedal tyo build
but it have those weird diodes ,is there any substituts for the values?
-
1N914 or 1N4148
-
So the only difference between the two MXR noise gates are the FET arrangement?? What's the big difference in sound/performance??
I didn't know there were 2 versions. I'm pretty sure I built the version 1. But I also followed the tonepad schematic. Dunno if the tonepad one was version 1 but I compared the schems and didn't remember seeing anything diffrent. But the version 2 schem posted here IS different.
Had no idea there were so many noise gates. That chinese one would be a good build if I made the PCB. The chinese schematic is hard to read, so I wouldn't be able to perf it. there's also this 4016 design:
(http://img315.imageshack.us/img315/2673/4016noisegate21sg.jpg)
But what I really wanna build is the DOD fx30 since I already have a working MXR noise gate built.
http://www.freeinfosociety.com/electronics/schemview.php?id=343
Anyone built this??
-
There is a certain level of complexity which you can't dip below, when it comes to noise control, unless you subcontract all the complexity out to a dedicated chip designed to do it. You will always need some attenuation element, whether FET or optoisolator, or OTA, and you will always need a rectifier to track the input signal, and you will always need some sort of gain stage to drive the rectifier. Everything beyond that is simply bells and whistles to make each element behave better
The Japanese schematic I posted is actually not all that complicated or exotic. The 2SC1000 trannies can be subbed with 2N5088s (watch out for the pinout differences), the 1S1555s can be 1N914s, and the LM324 can be a TL074 without any problems arising (though this would not improve performance in any way). The Zener is just any old 5.1V zener, and the K30A is reasonably available, though you can probably sub a 2N5457 for it if you need to.
The JFET is used as a voltage-controlled resistor, in parallel with the 470k fixed resistor. Between their combined parallel resistance, and the 10k resistorin the signal path just ahead of them, they form an attenuator/voltage-divider. When the JFET is turned on, the combined parallel resistance of the JFET and 470k drops, and they attenuate the signal. The extent to which the signal is attenuated is a function of the 10k and 470k resistors. If the 470k is decreased to, say 220k, then the drop in combined resistance produced when the JFET goes low is not that big a change from the resting state. Same thing if the 10k resistor is increased to 47k. If that "input leg" of our virtual pot has a greater resistance, then the "ground leg" formed by the 470k and JFET has to drop wayyyyyy down to produce a big change in attenuation. Conversely, if we change the 10k to 2k2, then it will take a huge change in JFET resistance to produce dramatic attenuation.
All of this is to say that you can tinker with the amount of attenuation produced by playing with either the value of the 10k, the value of the 470k, or both.
Much like the Dynacomp/Ross compressor, and all variants, the 1M Decay pot over on the right determines how long it takes to recharge that 10uf cap. Large series resistance between B+ and the + end of that cap means that current trickles in, rather than rushes in, so it takes longer to recharge the cap. The JFET is set to low resistance (i.e., the signal is "gated" off) when that cap is fully charged. So, the slower the recharge time, the longer the signal hangs around before the circuit cuts it out completely.
A gate with a variable decay time, and adjustable attenuation, will provide reasonably satisfactory performance over a fairly broad range of circumstances. If the threshold control is .twitchy", then let us know and some component-value adjustments can be made to improve the dialability.
Finally, please note that when I scanned the project article, I made no attempt to assure the scale. Plus, the layout is not flipped around in anticipation of PnP. You will likely have to do a screenshot of the layout, and then flip it over and resize, in whatever your graphics application of choice happens to be, before committing to a toner transfer.
-
here is one: http://hammer.ampage.org/files/LPM-23-NoiseGate.PDF
here is another: http://hammer.ampage.org/files/GainesNoiseGate.PDF
Those are both cool,I love the humor in the Japanese version,and the use of stars in the PCB.
Mark If I were to clean up the overlay and transfer to the Gains circuit would you look it over prior to me etching and wiring it up?
I am itching to etch something and need a noise gate.
Bryan
-
Here how does this look?
Cleaned up and mirrored,print scaling I am not sure about that may need a little work.
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=42722&g2_serialNumber=1)
-
Here's another one for the schematics collection... Haven't tried it but it does look pretty simple, very low parts count and one knob control...
Eingang = Input
Ausgang = Output
(http://moosapotamus.net/misc/noisgate1.gif)
I like the look of the Gaines NG, too. Full compliment of controls plus a key input. I could probably use a decent NG, myself. 8)
So, I whipped up an overlay, which usually helps me keep things in order when stuffing a board.
(http://moosapotamus.net/misc/GainesOVRsm.gif)
I'd probably replace the switch for the key input with a switching jack instead, just to avoid taking up space on the front panel. If I find time, I might even take a swing at redoing the layout to include a LT1054 (to handle the bipolar supply) and to make it fit vertically in a 1590BB enclosure.
~ Charlie
-
Like ANYthing with a sidechain, "feel" is important. And when it comes to sidechain-controlled devices, "feel" = time constants and the match of the time constants to the signal source and context. What makes for super gating of a kick drum, and helping to keep bleedthrough of amps to the drum mic when the drummer isn't hitting the kick drum, may be absolutely intrusive and interfering when it comes to a guitar.
Why am I telling you this? Because it is highly unlikely that a "simple" noise gate circuit will provide pleasing performance for a guitar unless it was deliberately designed for guitar from the ground up. So, in the circuit Charlie posted, will C3 provide the "right" decay characteristics? does the gain of T1 and the attenuation provided by R3 correspond to the range of signal levels a guitar presents? will T3/R5 provide the optimal amount of attenuation? We don't know.
This is not a scolding. Rather, like I say, whenever you're dealing with somethng whose functioning depends on the use of a detected signal envelope to manage a heard signal envelope, you cannot depend on any single detection circuit to yield the heard envelope you're looking for. The posted circuit may be absolutely what the doctor ordered for managing noisy backgrounds using a ham or CB radio carbon or crystal mic, but be entirely inappropriate for other uses.
As always, keep your expectations realistic.
-
Thanks for the caveats, Mark. 8)
I really only posted that one because the OP was looking for "simple". But I also stated that I had not built it so, yeah... [grain of salt].
Looks simple enough to breadboard relatively quickly, so it still might be worth a shot. Conceptually, I like the simplicity of the single control knob.
I am also thinking about building the Gates circuit, too.
~ Charlie
-
The DOD 230 is a really nice looking simple build and was designed for guitar as far as I know, i'm just about to build one up soon but from all reviews, seems to serve its purpose well and it's a nice easy optocoupler build so you just have to get an LDR that has a fast decay... seems to be pretty much that simple, no trim pots, anything and easily available values and just 2 Dual Op-Amps.
So although I haven't technically built it yet... it still gets my vote.
-
Like I keep saying, the dissatisfaction many have with noise gates is because they expect too much out of a single unit, obliging it to take care of too many "noise crimes". However, if you have a pair of them - one near the start and one near the end of your pedal chain - you can set them for gentler action such that they interfere with your sound less.
One of the great things about going DIY is that you can throw together more than one copy of a given circuit relatively inexpensively. This is one of those cases. :icon_wink:
-
I think I found a discrepancy between the schematic and PCB for the Gaines Noise Gate.
In the schematic, R9 and R27 are tied together. But in the PCB, instead of being tied together, it looks like they both go down to ground.
In the overlay I posted (above), R9 is near the bottom center of the layout, between C1 and R10. R27 is directly below IC2.
My gut says to modify the PCB and go with the schematic. Yes/no?
~ Charlie
-
I think I found a discrepancy between the schematic and PCB for the Gaines Noise Gate.
In the schematic, R9 and R27 are tied together. But in the PCB, instead of being tied together, it looks like they both go down to ground.
In the overlay I posted (above), R9 is near the bottom center of the layout, between C1 and R10. R27 is directly below IC2.
My gut says to modify the PCB and go with the schematic. Yes/no?
~ Charlie
Thanks for doing the overlay thats what I wanted to do ,but couldnt figure it out in Gimp.
And I think your right,I see what appears to be another problem based on what you already found.
I am going to evaluate it more and post back.
Thanks for catching that,Bryan
-
Here look at this, I think this will fix R9, R27, and C1 by tweaking a few traces and adding two jumpers.
But someone else should look it over >The red lines are where the new jumpers would go.
I also cleaned it up even more,so hopefully this will provide a good sharp transfer if its all correct.
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=42739&g2_serialNumber=1)
-
This one is easier to understand >
(http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=42742&g2_serialNumber=1)
-
Bryan - Do you have larger images? I think I see what you are doing, but It's kind of small. What was wrong with C1?
FYI - I am working on redoing the whole layout for a couple of reasons...
1) Needs a bipolar power supply, so I'm going to add a +/-15V supply that will run off of a standard +9VDC power supply.
2) The specified dual opamp is TL075 (that's not a typo). Obsolete and not pin for pin compatible with TL074, so rerouting.
3) Flipping the footprints for the pots around to allow PCB mounted pots that will orient up and fit, 4 across, in a 1590BB.
~ Charlie
-
Yeah I have the Gimp files if you want them,send me a PM with E-mail info.
Personally, I would rather just use what you draw up as you have WAY more experience at this than me ;)
I doubt I will use PCB mounted pots as I dont trust my precision, but I can just hand wire to the pads.
C1 was connected to R9 on the PCB and in the schematic it wasnt.Was I correct to unconnect it from R9 ?
Or was I looking at the PCB image wrong ?
thanks for the help,Bryan
-
Okay, I see. You got it right. C1 goes to ground, it doesn't connect to R9.
I think I've got the layout reworked to use a pair of dual opamps instead of the weird TL075 quad. Now I just need to squeeze in the bipolar power supply. I'll post it when it's done.
~ Charlie
-
The Q & D compressor 2 uses the SSM2166 ic (smallbear has them, about $8)
That ic has a downward expander inside. If you build the Q & D compressor with a compression ration of 1:1 (no compression) you have a noise gate.
Low parts count too..
http://www.muzique.com/ssm2166.htm (http://www.muzique.com/ssm2166.htm)
-
The Q & D compressor 2 uses the SSM2166 ic (smallbear has them, about $8)
That ic has a downward expander inside. If you build the Q & D compressor with a compression ration of 1:1 (no compression) you have a noise gate.
Low parts count too..
http://www.muzique.com/ssm2166.htm (http://www.muzique.com/ssm2166.htm)
The downward expansion is one of my favourite parts of the Q&D. Cures the perennial complaint of compressor-users: "Why is my compressor so noisy?". And while the chip itself is not exactly a 25-cent op-amp, so few additional parts are needed to make it useful that you can still have yourself a great compressor/noise-reduction system for under $15 (minus the chassis, stompswitch, etc.)
I know that HUSH chips were produced by SSM. Can they be purchased by "mere mortals" at all? http://www.zmitac.aei.polsl.pl/Electronics_Firm_Docs/Analog_pdf/ssm2000.pdf
-
Soon I'll be able to comment on the SSM2166, they are now somewhere over the ocean between the USA and Amsterdam.
I need to do some downward expansion after four PT2399's in series. The inter stage filtering is insufficient. I was really happy when I found this little chip. Haven't seen many downward only expanders, let alone simple ones.
I'll have to experiment with putting the expander in front and a filter at the end. Or the other way around.
You can see that it's prone to noise..though still surprisingly quiet :D
(http://www.newtone-online.nl/forum/index.php?action=dlattach;topic=4163.0;attach=4984;image)
-
I know that HUSH chips were produced by SSM. Can they be purchased by "mere mortals" at all? http://www.zmitac.aei.polsl.pl/Electronics_Firm_Docs/Analog_pdf/ssm2000.pdf
There is one on ebay for twentyfive bucks.. ::)
-
I might actually want a noise gate for drums. I'm a drummer and when I have my buddies over and we mic my kit there re some issues.The simple one posted by moosapotamus, would this work alright for that?
There's so many noise gates here now, I dunno which to commit to!
-
Arnoud,
The LM1894 is still available as far as I know, and is much cheaper than the SSM2166 and HUSH chips. It provides single-ended "noise-filtering", with 2 channels on a single DIP. http://www.nalanda.nitc.ac.in/industry/appnotes/Natsemi/AN-390.pdf
-
Thanks Mark,
I've seen that one too.. Interesting one as well. Needs little more signal conditioning at the front. Couldn't find supplier though. And I kinda like the fact that the SSM2166 can compress too. (found the ssm for about 4,95 euros now too ). Might try that one someday too.
-
I might actually want a noise gate for drums. I'm a drummer and when I have my buddies over and we mic my kit there re some issues.The simple one posted by moosapotamus, would this work alright for that?
There's so many noise gates here now, I dunno which to commit to!
I would not recommend "committing" to that simple one that I posted unless you breadboard it first and confirm that it works well, or can be tweaked to work well for you.
And then another option that appears to have gotten some rave reviews is the single-knobbed DOD230. I believe scheme and layouts have been posted in other threads here, as well as at FSB, so should be able to search it up pretty easily.
~ Charlie
-
DOD230 (http://www.diystompboxes.com/smfforum/index.php?topic=78349.msg645271)
-
http://i837.photobucket.com/albums/zz293/Nasse2/simplex1.jpg (http://i837.photobucket.com/albums/zz293/Nasse2/simplex1.jpg)
I was shovelling dust off of some of my "archives" and found one paper I have saved. Perhaps not on topic but perhaps it is useful for something. Never tested it. It might be nice on drums or something. Small component count but dual supply and perhaps not for everything
-
The DOD230? They made another noise gate? I know they had the DODfx30 and DODfx30b which are really different from each other...
great now MORE noise gates to choose from! That percussion noise gate would be cool.
-
For anyone interested in building the Gaines Noise Gate, I reworked the layout to replace the TL075 quad with a pair of TL072 duals. I also made two different power supply versions. The article references +/-15V for the circuit, so I did one layout with a 5V regulator and a 5-15V DC-DC converter to make +/-15V. But the DC-DC converter is a bit pricey and can sometimes be difficult to find. So I did another layout and replaced it with a LT1054 to supply +/-9V. My guess is that would give plenty of headroom for guitar playing. I have not built this yet, so both layouts are not verified.
(http://moosapotamus.net/misc/GainesNG15-OVR.gif)
(http://moosapotamus.net/misc/GainesNG9-OVR.gif)
The board dimensions are 3" x 1.8". Should fit nicely in a 1590BB.
Let me know if anyone finds any mistakes. Also if anyone wants to give it a go and wants a PCB transfer image, let me know and I'll post them, too.
~ Charlie
-
I gotta ask, what's "key in"?
-
Noise gates can be actuated by the input signal itself, OR by something else entirely different. That something different is the "key in".
So what good is it? Two examples:
1) If you listen to the David Bowie tune "Let's Dance", the horns sound sorta like horns, but sorta not. During production, producer Nile Rogers run the mixed-down horns through a gate, and used his guitar strumming to gate the horns in guitar-like fashion.
2) If a person wanted to stick a gate at the very end of the signal chain where the most noise had accumulated, but use the envelope of the original input signal, where the dynamics had the greatest contrast, they could run the guitar into a splitter. One output from the splitter would go to the rest of the pedal chain, and the other would go to the key in, so that the original guitar dynamics could be used for controlling the gate.
-
Just to add to Mark's description... The key input is the input to the envelope trigger for the noise gate. When you use the key input, it becomes much less about noise gating and more about one input signal triggering (opening the gate) for another input signal. Another application example... split your guitar signal to your amp and to the key input, then plug an AM radio station into the noise gate input. Then whenever you play the guitar, the gate opens and you will also hear the AM radio. Stop playing guitar and the gate shuts and the AM radio gets muted, too.
~ Charlie
-
you guys are awesome. this place is awesome!
-
I reckon! Best diy forum, surely? :)
-
Thanks to everyone that has helped me out with this.
I decided on building the MXR noise gate due to it's sound and it's not that difficult.
I ordered all the parts and will start building it soon. :]
-
Well I dunno how to etch PCB boards, so I'm going with perf. I think I'm gunna do the DODfx30.Eventually I'll do the gains tho. And another MXR noise gate too. And then work on something for my drum mics.
-
I am waiting for Charlies redesign of the Gains board and will build it...
But good luck on your builds let us know how they come out,Bryan
-
I would learn PCB to do this.
-
I am waiting for Charlies redesign of the Gains board and will build it...
But good luck on your builds let us know how they come out,Bryan
Hey Bryan - I posted my redesign of the Gates NG a few posts above...
http://www.diystompboxes.com/smfforum/index.php?topic=81918.msg697345#msg697345
... two versions, 1) with +/-15V PSU, 2) with +/-9V PSU.
Also replaced the odd TL075 quad opamp with a pair of duals (TL072 or similar).
The parts list is the same as in the Gains article, plus the additional parts for the PSU. Instead of part numbers, the actual values of the additional parts for the PSU are indicated in the PCB layouts.
If you are waiting for it to be verified, I'm afraid that I have a couple of other projects to finish before I get to this one. But if you want to give it a go, I'd be glad to make a couple of PCB transfer images and post them, too. LMK
El - theres a pretty good photo essay on making PCBs at Tonepad...
http://tonepad.com/photoessay.asp?photoEssayID=10&sequenceNo=1
... definitely check it out.
~ Charlie
-
Thanks Chalie IDK how I missed that :icon_redface:
yeah can you post the transfers I definetly want to etch mine.
Now that I have all the stuff its actually cheaper to etch than use perf,and works much better to boot.
El > also look at John Lyons web site he has a very good tut on etching on the cheap.
AND PLEASE dont flush your etchent like Tonepad recomends !!!!! I cant believe he wrote that...
The heavy amounts of copper is VEY BAD for the eco system ,please,please dispose of properly if you do start etching.
-
Here's the corresponding toner transfer images. Resize the image so the actual board dimensions are 3" x 1.8".
(http://moosapotamus.net/misc/GainesNG15-PCB.gif)
(http://moosapotamus.net/misc/GainesNG9-PCB.gif)
Both are currently UNVERIFIED. Please post if you build it.
~ Charlie
-
Thank you for sharing Charlie 8) 8) 8)
Yeah once I get it done I will post a report and some pics,I am etching for upcoming builds this week so I will throw these in the list.
Again thanks so much for doing a new layout and sharing your work,I REALLY appreciate that. :icon_mrgreen:
Bryan
-
I'm new here and hope it's ok to respond more than a year later. I'm hoping the the folks discussing this are still participating in the forum.
I just built the Gaines noise gate, but didn't see this forum until I was finished. I wanted to comment on the R9/R27 issue and hopefully get some troubleshooting help.
First, I tried the circuit with the junction of R9 and R27 grounded as shown on the PCB layout, and then ungrounded as shown on the schematic. I found that when I lifted the ground, the volume went way down and the LED barely lit up. I decided that maybe it should be grounded and Gaines forgot to update the schematic.
My main problem is that the unit distorts the guitar signal when the gate is on. I tried increasing the value of C1 to 4.7uf, and that helped the sputtering sound as the gate closes, but not the distortion while the gate is open. Has anyone else encountered this? Any suggestions for cleaning up the signal?
-
My main problem is that the unit distorts the guitar signal when the gate is on. I tried increasing the value of C1 to 4.7uf, and that helped the sputtering sound as the gate closes, but not the distortion while the gate is open. Has anyone else encountered this? Any suggestions for cleaning up the signal?
Have anyone else built this pedal? Any other distortion experiences? :icon_question:
-
The mxr noise gate project posted at tonepad works really well.
It has a bad reputation due to some isolated cases where some people couldn't get it to work, my personal suggestion is use a socket for the FET and try some in there, make sure the pinout is right too.
-
I've found another thread about Gaines Noise Gate here: http://www.diystompboxes.com/smfforum/index.php?topic=103581.msg1044952#msg1044952
It seems that this schematic contains an error, e.g. R27 and R9 joint shall be connected to pin 1 of threshold pot.
-
Is using 741 crucial, or I can use TL071 there?
-
I have finally modelled this with LTspice, and it is quite obvious that original schematic is good - R27 and R9 should be connected only to each other. I would like to share LTspice file for this, but I don't know how to attach a file here besides picture. I have added input buffer to provide necessary amplification, because this gate works on studio signal level.
(https://s18.postimg.org/rmpjeg1gl/gaines.jpg) (https://postimg.org/image/rmpjeg1gl/)
However, I am still having trouble to make it running in reality. :(
-
I am waiting for Charlies redesign of the Gains board and will build it...
But good luck on your builds let us know how they come out,Bryan
Hey Bryan - I posted my redesign of the Gates NG a few posts above...
http://www.diystompboxes.com/smfforum/index.php?topic=81918.msg697345#msg697345
... two versions, 1) with +/-15V PSU, 2) with +/-9V PSU.
Also replaced the odd TL075 quad opamp with a pair of duals (TL072 or similar).
The parts list is the same as in the Gains article, plus the additional parts for the PSU. Instead of part numbers, the actual values of the additional parts for the PSU are indicated in the PCB layouts.
~ Charlie
It's been a few years, can anyone share this
-
I am waiting for Charlies redesign of the Gains board and will build it...
But good luck on your builds let us know how they come out,Bryan
Hey Bryan - I posted my redesign of the Gates NG a few posts above...
http://www.diystompboxes.com/smfforum/index.php?topic=81918.msg697345#msg697345
... two versions, 1) with +/-15V PSU, 2) with +/-9V PSU.
Also replaced the odd TL075 quad opamp with a pair of duals (TL072 or similar).
The parts list is the same as in the Gains article, plus the additional parts for the PSU. Instead of part numbers, the actual values of the additional parts for the PSU are indicated in the PCB layouts.
~ Charlie
It's been a few years, can anyone share this
If you need original article, here it is: https://docdro.id/GlDO0AQ