Jump to content

Guys with knowledge of circuit design, give me a hand.


ej

Recommended Posts

  • Members

I designed this simple 1 transistor distortion with a switchable band pass filter for lowfi sounds. I would love some feedback on it.

FilterFuzz_zps2c74386c.jpg

This is my first time trying my hand at anything like this. I left out all the switching and such. What do y'all think?

Link to comment
Share on other sites

  • Members

Either develop a passion for breadboards or use a simulator. A good sim will teach you an awful lot about how changes affect your circuit. You can use the simulator on circuitlab.com for free, and you don't have to download anything because it runs in your web browser. The downside is that it's all java script so it's really slow to run a simulation. I tried simulating only the octave amp of an FY-6 on circuitlab and it took more than an hour to generate 1000 sample points.

There are lots of free Spice simulation engines but few with graphical interfaces. Writing simulation scripts manually isn't my idea of a pleasant afternoon, but I've done it before.

I'm currently using 5Spice (5spice.com). The free demo version is very limited - very small schematic drawing space, and some of the simulation features are disabled, but it will give you an idea how it works. Once you register it (around $200) then it unlocks all the features, and you can draw very large schematics to simulate. I can simulate the entire six transistor FY-6 circuit and collect 10,000 sample points in a little over a second. It's very accurate, too. The waveforms I saw in simulation were exactly what I saw on the bench with the breadboard and o-scope.

You'll need Spice models for every component you use in your circuit. 5Spice comes with some common components, and many more can be found on the internet. They're just text files containing a description of how the component performs, but they contain parameters you often won't find on a manufacturer's datasheet. For semiconductors, you can usually find Spice models on the manufacturer's website.

Link to comment
Share on other sites

  • Members

Either develop a passion for breadboards or use a simulator. A good sim will teach you an awful lot about how changes affect your circuit. You can use the simulator on circuitlab.com for free, and you don't have to download anything because it runs in your web browser. The downside is that it's all java script so it's really slow to run a simulation. I tried simulating only the octave amp of an FY-6 on circuitlab and it took more than an hour to generate 1000 sample points.

There are lots of free Spice simulation engines but few with graphical interfaces. Writing simulation scripts manually isn't my idea of a pleasant afternoon, but I've done it before.

I'm currently using 5Spice (5spice.com). The free demo version is very limited - very small schematic drawing space, and some of the simulation features are disabled, but it will give you an idea how it works. Once you register it (around $200) then it unlocks all the features, and you can draw very large schematics to simulate. I can simulate the entire six transistor FY-6 circuit and collect 10,000 sample points in a little over a second. It's very accurate, too. The waveforms I saw in simulation were exactly what I saw on the bench with the breadboard and o-scope.

You'll need Spice models for every component you use in your circuit. 5Spice comes with some common components, and many more can be found on the internet. They're just text files containing a description of how the component performs, but they contain parameters you often won't find on a manufacturer's datasheet. For semiconductors, you can usually find Spice models on the manufacturer's website.

Link to comment
Share on other sites

  • Members
Quote Originally Posted by amp_surgeon View Post
Either develop a passion for breadboards or use a simulator. A good sim will teach you an awful lot about how changes affect your circuit. You can use the simulator on circuitlab.com for free, and you don't have to download anything because it runs in your web browser. The downside is that it's all java script so it's really slow to run a simulation. I tried simulating only the octave amp of an FY-6 on circuitlab and it took more than an hour to generate 1000 sample points.

There are lots of free Spice simulation engines but few with graphical interfaces. Writing simulation scripts manually isn't my idea of a pleasant afternoon, but I've done it before.

I'm currently using 5Spice (5spice.com). The free demo version is very limited - very small schematic drawing space, and some of the simulation features are disabled, but it will give you an idea how it works. Once you register it (around $200) then it unlocks all the features, and you can draw very large schematics to simulate. I can simulate the entire six transistor FY-6 circuit and collect 10,000 sample points in a little over a second. It's very accurate, too. The waveforms I saw in simulation were exactly what I saw on the bench with the breadboard and o-scope.

You'll need Spice models for every component you use in your circuit. 5Spice comes with some common components, and many more can be found on the internet. They're just text files containing a description of how the component performs, but they contain parameters you often won't find on a manufacturer's datasheet. For semiconductors, you can usually find Spice models on the manufacturer's website.
Thanks dude! I was actually playing around with CircuitLab this morning. Its currently trying to generate the output waveform, should be another 20 minutes facepalm.gif

Im gonna grab the free version of 5Spice.
Link to comment
Share on other sites

  • Members
Quote Originally Posted by amp_surgeon View Post
Either develop a passion for breadboards or use a simulator. A good sim will teach you an awful lot about how changes affect your circuit. You can use the simulator on circuitlab.com for free, and you don't have to download anything because it runs in your web browser. The downside is that it's all java script so it's really slow to run a simulation. I tried simulating only the octave amp of an FY-6 on circuitlab and it took more than an hour to generate 1000 sample points.

There are lots of free Spice simulation engines but few with graphical interfaces. Writing simulation scripts manually isn't my idea of a pleasant afternoon, but I've done it before.

I'm currently using 5Spice (5spice.com). The free demo version is very limited - very small schematic drawing space, and some of the simulation features are disabled, but it will give you an idea how it works. Once you register it (around $200) then it unlocks all the features, and you can draw very large schematics to simulate. I can simulate the entire six transistor FY-6 circuit and collect 10,000 sample points in a little over a second. It's very accurate, too. The waveforms I saw in simulation were exactly what I saw on the bench with the breadboard and o-scope.

You'll need Spice models for every component you use in your circuit. 5Spice comes with some common components, and many more can be found on the internet. They're just text files containing a description of how the component performs, but they contain parameters you often won't find on a manufacturer's datasheet. For semiconductors, you can usually find Spice models on the manufacturer's website.
Thanks dude! I was actually playing around with CircuitLab this morning. Its currently trying to generate the output waveform, should be another 20 minutes facepalm.gif

Im gonna grab the free version of 5Spice.
Link to comment
Share on other sites

  • Members

So I spent the day tweaking the circuit in CircuitLab. Heres what I have:

FuzzFilter_zps8683147f.png

FuzzFilterFreq_zps3b68c5e7.png

FilterFuzzoutput_zpsd08df5f2.png

What do y'all think? Theres a significant drop in output, but that's to be expected when you're cutting out a huge range of frequencies.

Link to comment
Share on other sites

  • Members

So I spent the day tweaking the circuit in CircuitLab. Heres what I have:

FuzzFilter_zps8683147f.png

FuzzFilterFreq_zps3b68c5e7.png

FilterFuzzoutput_zpsd08df5f2.png

What do y'all think? Theres a significant drop in output, but that's to be expected when you're cutting out a huge range of frequencies.

Link to comment
Share on other sites

  • Members
Quote Originally Posted by ej View Post
So I spent the day tweaking the circuit in CircuitLab. Heres what I have:

FuzzFilter_zps8683147f.png

FuzzFilterFreq_zps3b68c5e7.png

FilterFuzzoutput_zpsd08df5f2.png

What do y'all think? Theres a significant drop in output, but that's to be expected when you're cutting out a huge range of frequencies.
Where is R25 set? I seem to recall the default is 50%. Also, with emitter resistors that small you're going to be clipping in both transistor stages even with a pretty low level signal. Did you mean for this to clip in the transistor stages, or were you looking for the diodes to do the clipping?

And seriously, 2Vpp is a pretty hot input, and 1kHz is pretty high frequency. Try it at 500mV and 100Hz or 500Hz.
Link to comment
Share on other sites

  • Members
Quote Originally Posted by ej View Post
So I spent the day tweaking the circuit in CircuitLab. Heres what I have:

FuzzFilter_zps8683147f.png

FuzzFilterFreq_zps3b68c5e7.png

FilterFuzzoutput_zpsd08df5f2.png

What do y'all think? Theres a significant drop in output, but that's to be expected when you're cutting out a huge range of frequencies.
Where is R25 set? I seem to recall the default is 50%. Also, with emitter resistors that small you're going to be clipping in both transistor stages even with a pretty low level signal. Did you mean for this to clip in the transistor stages, or were you looking for the diodes to do the clipping?

And seriously, 2Vpp is a pretty hot input, and 1kHz is pretty high frequency. Try it at 500mV and 100Hz or 500Hz.
Link to comment
Share on other sites

Quote Originally Posted by ej View Post

What do y'all think? Theres a significant drop in output, but that's to be expected when you're cutting out a huge range of frequencies.
I think someone already recommended a make-up gain stage to compensate for this. IMHO, it would be a good idea if you're going to use a fairly drastic bandpass filter on the signal.
Link to comment
Share on other sites

  • Members
Quote Originally Posted by ej View Post
So I spent the day tweaking the circuit in CircuitLab. Heres what I have:

FuzzFilter_zps8683147f.png

[...]

What do y'all think? Theres a significant drop in output, but that's to be expected when you're cutting out a huge range of frequencies.
First of all this thread is cool, because of subject. But this has got problems. Also you're overdriving it hugely with that input level.

The lack of a coupling cap distorts out D1 and D2 from Q4's bias.
The bias on both stages is probably almost cutoff. Look at the collector voltages.
If R25 is 100K etc. then in needs to be after the network. This would also eliminate C14, except reuse it on the first item.
Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...