diode bridge rectifier as a clipper?

[HydrozeenElectronics]

I've been using 1n34a and when i hook up the battery and start playing they sound one way then after a few minutes of having power turned on to them they start to sound another way as if they "warmed up"

I have used them in different build and always had that same effect of warming up.

any ideas on what that might be?

I had an idea to make the GE diodes stable I was thinking of placing a small flashlight size bulb next to them to act as a heater. I would have to have external power but I was thinking if I got the temp up to like 99 degrees then in anything but a smoldering heat wave I should get reliable consistency out of them. Does anyone see any pitfalls to that? would that shorten the life of the diodes?

Thanks

I've actually had no temperature instability problems with Ge diodes at all,they have all worked fine for me,I think the only way to shorten the life/destroy a Ge diode,is to exceede their power rating,or their PIV (Peak Inverse Voltage)....

[MoonWatcher]

Actually, in the bridge rectifer configuration in the Tonebone/Marshall JCM-1 info linked by lolbou, you end up with three diodes in series in each direction, not two.

There's actually a better way - replace the 5th diode with a simple jumper. It's what is in the Timmy pedal (albeit in the negative feedback loop), but it has a definite effect that I like better than the extra diode.

I found out about that trick by accident. Back when it was "okay" to have a pedal in a big honkin' box, I'd use a bridge simply because it was a simple way to do either a pair of diodes, or a quad in the popular series/parallel configuration. I would use a DPDT to short out a pair of them to ground, to give me the two "modes." Accidentally, I didn't connect the switch to ground in one instance, so it simply jumpered the series pairs at their junctions only.

I really like it in a neg feedback loop though - Paul C. must have figured out the same thing.

I'd love a technical explanation on what's happening too, since I can't wrap my head around it. I bet a 'scope would really help to see the differences, but I'm currently without one.

[HydrozeenElectronics]

Seems to me to get really good Harmonic Tone you need some kind of crossover distortion. I the bridge rectifier is not just shunting the signal to ground but re-injecting a little bit of the signal slightly out of phase and causing some cross over distortion much like a Fuzzrite distortion.

I think that's also why people say 100 watt marshalls are more "complex" sounding because with 4 tubes pushing and pulling instead of two you get a little more cross over distortion.

[mictester]

Update. I just built a rectifier circuit based on the one in the marshall schematic out of Germanium Diodes. It sounds pretty damn good. I a/b'ed with a simple germanium diode clipping pair shunted to ground and the rectifier type has more harmonics and is louder.
Cheers

There won't be any "more harmonics" unless you're driving harder into something later in the circuit which is giving another layer of clipping. It WILL be louder because you've just doubled the diode drop you're seeing, by putting extra diodes in series!

A more interesting approach is to use the diodes as a clipper, but also put a variable DC current through the diodes, so you can also use them as a VCA (that's voltage-controlled attenuator) for compression purposes. The closest I've ever got to "valve sound" is using this kind of configuration.

[DrNomis]

I've been using 1n34a and when i hook up the battery and start playing they sound one way then after a few minutes of having power turned on to them they start to sound another way as if they "warmed up"

I have used them in different build and always had that same effect of warming up.

any ideas on what that might be?

I had an idea to make the GE diodes stable I was thinking of placing a small flashlight size bulb next to them to act as a heater. I would have to have external power but I was thinking if I got the temp up to like 99 degrees then in anything but a smoldering heat wave I should get reliable consistency out of them. Does anyone see any pitfalls to that? would that shorten the life of the diodes?

Thanks

I've actually had no temperature instability problems with Ge diodes at all,they have all worked fine for me,I think the only way to shorten the life/destroy a Ge diode,is to exceede their power rating,or their PIV (Peak Inverse Voltage)....

It may be that there is a bit of a drift in the characteristics of the diodes with temperature,they are Germanium types after all,normally you see temperature drifts in active Germanium devices,it would be hard to tell exactly what's causing the "warming-up" that you are noticing,without seeing the schematic...

The clipping produced by a couple of back-to-back Ge diodes is softer than Si diodes,that is,the corners are more rounded,as you may well know....

Shottky Diodes give similar clipping to Ge diodes...

[earthtonesaudio]

Shottky Diodes give similar clipping to Ge diodes...

The forward voltage is similar but other parameters are very different.

[RnFR]

yup. one important one when talking about clippers is the conduction knee is softer.

[DrNomis]

yup. one important one when talking about clippers is the conduction knee is softer.

My 20Mhz Dual Trace Scope has a built-in component tester,which is perfect if we need some screenshots of what the clipping characteristics are like,would you like me to post some screenshots RnFR,it's easily done....

[RnFR]

sure, if you want to go ahead. but, I've seen the curves before. they are subtle, but there are differences. I believe Merlin blencowe posted them recently.

I wouldn't mind seeing a comprehensive set of curves for a number of diodes that we use though. I think it would be interesting.

[DrNomis]

sure, if you want to go ahead. but, I've seen the curves before. they are subtle, but there are differences. I believe Merlin blencowe posted them recently.

I wouldn't mind seeing a comprehensive set of curves for a number of diodes that we use though. I think it would be interesting.

No worries,well,since this forum is about the bridge-rectifier-clipper in the JCM900,I'll start the ball rolling with it,I think I have some germanium diodes floating around,and Bajaman was curious about a Led version,so I'll do a Led version after the bridge rectifier,might take a while,so stay tuned...

[RnFR]

sounds cool!

[DrNomis]

Okay,here is what I saw on the screen with the JCM900 Bridge-Rectifier-Clipper...


Note,you may need to rotate the image in your mind....


Bridge Rectifier used for this demo: KBPC602 with a UF4003 soldered between the + and - terminals(Anode to the +,and Cathode to the -).

The UF4003 is an Ultra-Fast version of the 1N4003 power diode...

[DrNomis]

Okay,here we have it,this is what I saw on the screen with the same circuit as before,implemented in Leds...

I used 5 Red 5mm Leds to build the circuit on a solderless breadboard with wires going to the component tester on my scope....


As a sidenote,when I connected the circuit up to the component tester and pressed the component test button,all the leds lit up....

[DrNomis]

First of all,what do we notice when we compare the two screenshots?....

Well,we can see that the clipping sharpness is deffinitely different,there is slightly more rounding in the Led Clipper,but,that's not all,notice how the two bends in the line are further apart?,yep,that is because we now have a larger voltage drop across each Led,a Led requires at least 1.5V across it before it turns on,this is more than the voltage drop across a silicon power diode,the KBPC602 bridge rectifier contains 4 Silicon diodes,each one requiring .6V across it to turn on,with that extra UF4003 diode,each path taken by the currents in the circuit,goes through 3 Si diodes,a total of .6V+.6V+.6V=1.8V For three Silicon diodes,compared with 1.5+1.5V+1.5V=4.5V total for three Leds,what can we conclude from this?....

Firstly,if we use a Bridge Rectifier circuit implemented by Leds,the signal will be clipped later,but our signal level will be greater....


Please feel free to check for any errors in my maths and point it out for me if you do find any....

[HydrozeenElectronics]

Very cool, thanks Could you build a Germanium Rectifier and give a screen shot of it as well?


Cheers

First of all,what do we notice when we compare the two screenshots?....

Well,we can see that the clipping sharpness is deffinitely different,there is slightly more rounding in the Led Clipper,but,that's not all,notice how the two bends in the line are further apart?,yep,that is because we now have a larger voltage drop across each Led,a Led requires at least 1.5V across it before it turns on,this is more than the voltage drop across a silicon power diode,the KBPC602 bridge rectifier contains 4 Silicon diodes,each one requiring .6V across it to turn on,with that extra UF4003 diode,each path taken by the currents in the circuit,goes through 3 Si diodes,a total of .6V+.6V+.6V=1.8V For three Silicon diodes,compared with 1.5+1.5V+1.5V=4.5V total for three Leds,what can we conclude from this?....

Firstly,if we use a Bridge Rectifier circuit implemented by Leds,the signal will be clipped later,but our signal level will be greater....


Please feel free to check for any errors in my maths and point it out for me if you do find any....

[DrNomis]

Very cool, thanks Could you build a Germanium Rectifier and give a screen shot of it as well?


Cheers

First of all,what do we notice when we compare the two screenshots?....

Well,we can see that the clipping sharpness is deffinitely different,there is slightly more rounding in the Led Clipper,but,that's not all,notice how the two bends in the line are further apart?,yep,that is because we now have a larger voltage drop across each Led,a Led requires at least 1.5V across it before it turns on,this is more than the voltage drop across a silicon power diode,the KBPC602 bridge rectifier contains 4 Silicon diodes,each one requiring .6V across it to turn on,with that extra UF4003 diode,each path taken by the currents in the circuit,goes through 3 Si diodes,a total of .6V+.6V+.6V=1.8V For three Silicon diodes,compared with 1.5+1.5V+1.5V=4.5V total for three Leds,what can we conclude from this?....

Firstly,if we use a Bridge Rectifier circuit implemented by Leds,the signal will be clipped later,but our signal level will be greater....


Please feel free to check for any errors in my maths and point it out for me if you do find any....

I could,but I'm fresh out of Ge diodes,I'll have to buy a few first,but yeah,I'm also keen to see if there is much difference to the clipping if Ge diodes are used,I'm speculating that the clipping will be quite soft,with rounded corners.....

I'll tell you what,since I need to buy some anyway,I'll buy some from my local electronics parts source and get it all sorted out this afternoon,if you're willing to wait for the results,as I feel this needs investigating,okay so stay tuned.....

[HydrozeenElectronics]

That would be no problem waiting. I'll be looking forward to it.

thanks

Very cool, thanks Could you build a Germanium Rectifier and give a screen shot of it as well?


Cheers

First of all,what do we notice when we compare the two screenshots?....

Well,we can see that the clipping sharpness is deffinitely different,there is slightly more rounding in the Led Clipper,but,that's not all,notice how the two bends in the line are further apart?,yep,that is because we now have a larger voltage drop across each Led,a Led requires at least 1.5V across it before it turns on,this is more than the voltage drop across a silicon power diode,the KBPC602 bridge rectifier contains 4 Silicon diodes,each one requiring .6V across it to turn on,with that extra UF4003 diode,each path taken by the currents in the circuit,goes through 3 Si diodes,a total of .6V+.6V+.6V=1.8V For three Silicon diodes,compared with 1.5+1.5V+1.5V=4.5V total for three Leds,what can we conclude from this?....

Firstly,if we use a Bridge Rectifier circuit implemented by Leds,the signal will be clipped later,but our signal level will be greater....


Please feel free to check for any errors in my maths and point it out for me if you do find any....

I could,but I'm fresh out of Ge diodes,I'll have to buy a few first,but yeah,I'm also keen to see if there is much difference to the clipping if Ge diodes are used,I'm speculating that the clipping will be quite soft,with rounded corners.....

I'll tell you what,since I need to buy some anyway,I'll buy some from my local electronics parts source and get it all sorted out this afternoon,if you're willing to wait for the results,as I feel this needs investigating,okay so stay tuned.....

[HydrozeenElectronics]

I tried that this afternoon. It was quite interesting. Using 5 GE diodes the sound was very Marshally' for lack of a better workd and when I used a jumper it took on a Mesa type overdrive sound. I think this would be really cool to have on a switch in a pedal.

Thanks for the tip.

Cheers

Actually, in the bridge rectifer configuration in the Tonebone/Marshall JCM-1 info linked by lolbou, you end up with three diodes in series in each direction, not two.

There's actually a better way - replace the 5th diode with a simple jumper. It's what is in the Timmy pedal (albeit in the negative feedback loop), but it has a definite effect that I like better than the extra diode.

I found out about that trick by accident. Back when it was "okay" to have a pedal in a big honkin' box, I'd use a bridge simply because it was a simple way to do either a pair of diodes, or a quad in the popular series/parallel configuration. I would use a DPDT to short out a pair of them to ground, to give me the two "modes." Accidentally, I didn't connect the switch to ground in one instance, so it simply jumpered the series pairs at their junctions only.

I really like it in a neg feedback loop though - Paul C. must have figured out the same thing.

I'd love a technical explanation on what's happening too, since I can't wrap my head around it. I bet a 'scope would really help to see the differences, but I'm currently without one.

[DrNomis]

I tried that this afternoon. It was quite interesting. Using 5 GE diodes the sound was very Marshally' for lack of a better workd and when I used a jumper it took on a Mesa type overdrive sound. I think this would be really cool to have on a switch in a pedal.

Thanks for the tip.

Cheers

Actually, in the bridge rectifer configuration in the Tonebone/Marshall JCM-1 info linked by lolbou, you end up with three diodes in series in each direction, not two.

There's actually a better way - replace the 5th diode with a simple jumper. It's what is in the Timmy pedal (albeit in the negative feedback loop), but it has a definite effect that I like better than the extra diode.

I found out about that trick by accident. Back when it was "okay" to have a pedal in a big honkin' box, I'd use a bridge simply because it was a simple way to do either a pair of diodes, or a quad in the popular series/parallel configuration. I would use a DPDT to short out a pair of them to ground, to give me the two "modes." Accidentally, I didn't connect the switch to ground in one instance, so it simply jumpered the series pairs at their junctions only.

I really like it in a neg feedback loop though - Paul C. must have figured out the same thing.

I'd love a technical explanation on what's happening too, since I can't wrap my head around it. I bet a 'scope would really help to see the differences, but I'm currently without one.

Hmmmm...very interesting,so,you could theoreticaly build a stompbox which has a feature enabling you to switch between Marshall and Mesa/Boogie sounds "on-the-fly" in the middle of a song......

[DrNomis]

Okay,I've just hit a bit of a snag,Jaycar Electronics,the only parts stockist here in Darwin,don't seem to stock any Ge Diodes in their parts inventory,so,what to do?...Hmmm....