1KHz 0dBu pocket generator

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Here is the schematic of the simple TL072 version. I need to add a switch for the battery and a cap across it to lower its ac impedance. I have arranged it so the output impedance is about 150 ohms and added a 0dBu / -40dBu switch for testing both line and mic inputs.

Cheers

Ian
 

Attachments

  • 1KOSC.png
    1KOSC.png
    43.9 KB · Views: 77
Here is the first draft of the PCB. The board is 50mm square:

pocketoscv1.png


J2 is the battery input and S1 is the on/off switch. The switch can be mounted on the PCB or remotely. J1 is the output. I left it as a 3 pin Molex so you can attach it to any XLR or TRS you like. S2 selects either 0dBu or -40dBu output. TP21 and 0VA allow you to set the output to 0dBu using VR1 which sets the gain of the second stage. J3 connect to pin 1 of the Molex in case you use a plastic XLR and a metal front panel so you can connect the panel to pin1.

Cheers

Ian
 
The 6.8K and 3.3K resistors make up a gain of around 3.06, not enough to start oscillations, you should try lowering the 3.3K resistor to 3K or even 2.7K and add a resistor in series with the diodes to lower and stabilize the gain once it starts oscillating.

I would go with a 22k or 24K and a 10K for a gain of 3.2-3.4, plus a 56K-100K in series with the diodes
 
user 37518 said:
The 6.8K and 3.3K resistors make up a gain of around 3.06, not enough to start oscillations, you should try lowering the 3.3K resistor to 3K or even 2.7K and add a resistor in series with the diodes to lower and stabilize the gain once it starts oscillating.

I would go with a 22k or 24K and a 10K for a gain of 3.2-3.4, plus a 56K-100K in series with the diodes
The gain required for oscillation is 3.0. The breadboard does burst into oscillation. The Wien bridge element is a direct copy of the one of an existing design posted earlier in the thread. However, I was concerned that even with 1% tolerance resistors, there would be occasions where the gain was less than 3.0 and oscillation would not start. I did once try a 3K but I think the output clipped.

The purpose of the diodes appears to be to stabilise the output amplitude rather than to stabilise the gain.

Cheers

Ian
 
ruffrecords said:
The gain required for oscillation is 3.0. The breadboard does burst into oscillation. The Wien bridge element is a direct copy of the one of an existing design posted earlier in the thread. However, I was concerned that even with 1% tolerance resistors, there would be occasions where the gain was less than 3.0 and oscillation would not start. I did once try a 3K but I think the output clipped.

The purpose of the diodes appears to be to stabilise the output amplitude rather than to stabilise the gain.

Cheers

Ian

Ian, the gain of 3 is the textbook gain to satisfy Barkaussen criterion to sustain oscillations, however, many times you need a gain of higher than 3 to guarantee oscillations every time the circuit starts, you need to move the poles to the right so the oscillations start growing, then as oscillations grow, the diodes conduct and if you place a resistor in series with them this resistor will be in parallel with the feedback resistor as the diodes begin to conduct, thus restoring gain to 3, the diode/resistor combination works as an AGC to avoid saturation. I've built these circuits dozens of times and they sometimes start oscillating with a gain of exactly 3, many times they do not. Check out Sedra/Smith's book or similar, I can give you plenty book examples like the one I am refering to.
 
user 37518 said:
Ian, the gain of 3 is the textbook gain to satisfy Barkaussen criterion to sustain oscillations, however, many times you need a gain of higher than 3 to guarantee oscillations every time the circuit starts, you need to move the poles to the right so the oscillations start growing, then as oscillations grow, the diodes conduct and if you place a resistor in series with them this resistor will be in parallel with the feedback resistor as the diodes begin to conduct, thus restoring gain to 3, the diode/resistor combination works as an AGC to avoid saturation. I've built these circuits dozens of times and they sometimes start oscillating with a gain of exactly 3, many times they do not. Check out Sedra/Smith's book or similar, I can give you plenty book examples like the one I am refering to.

Yes, books refs would be good. I am always willing to learn.

Cheers

Iamn
 
I have updated both the schematic and PCB layout based on the comments posted here. There is now a trimpot for waveform as well as level so you can minimise distortion and calibrate the output level. A switch selects either 0dBu or -40dBu.  On the PCB I have include a 0VA and output test points for easy set up on the bench. The board is 50mm square. Schematic attached.

I will let the design fester for a few days and then order a bunch of PCBs. If anyone wants one let me know so I can adjust the quantities appropriately.

Cheers

Ian
 

Attachments

  • 1KOSC.png
    1KOSC.png
    46.7 KB · Views: 50
Hi Ian,
if its not too late you might pencil me in for one of the boards .

You might let me know the price & postage , also is the pinout/layout/switch positions compatible with the Behringer CT-100 enclosure ?
 
Tubetec said:
Hi Ian,
if its not too late you might pencil me in for one of the boards .

You might let me know the price & postage , also is the pinout/layout/switch positions compatible with the Behringer CT-100 enclosure ?

You are pencilled in. I will only be charging postage to groupDIY members. I am getting the boards for next to nothing from China.. Then I will sell them on eBay and advertise them on GS!

They are net designed specifically to be compatible with the CT-100 box. I took mine apart and discovered it uses a PIC.

Cheers

Ian
 
I just picked up this Ivie IE-20B pocket white/pink noise generator for $5.00 on ebay:
https://www.ivie.com/download/IE-20B.pdf

The NiCad batteries were shot, but there is plenty of room for a 9V and a TL7805.  It goes from 0dB (adjustable to match *your* zero level) to -58dB in 2dB steps.  The only possible complaint is the RCA output.
 
mjrippe said:
I just picked up this Ivie IE-20B pocket white/pink noise generator for $5.00 on ebay:
https://www.ivie.com/download/IE-20B.pdf

The NiCad batteries were shot, but there is plenty of room for a 9V and a TL7805.  It goes from 0dB (adjustable to match *your* zero level) to -58dB in 2dB steps.  The only possible complaint is the RCA output.

Interesting. Looks like it uses a filtered pseudo random number generator to create white/pink noise. COuld probably be done with a small PIC micro. Thanks for posting.

Cheers

Ian
 
Hi Ian,

I’ll take a couple of boards if I may. One for me, and I have a client in mind who would find one very handy.

Let me know cost and postage.

Peter
 
I will keep you posted. The boards arrived the other day. I have built one up and tested it. dc voltages all look OK but it is not oscillating. Got some debugging to do.

Cheers

Ian
 
I found the fault on the PCB. The positive feedback network was routed to pin 2 instead of pin 3. Fortunately there is an easy to cut track to disconnect it. So I cut it and added a wire link to pin 3. Plugged it in and hey presto - the ouput slowly creeps down the the -ve rail. Duh?? The I look again - forgot to plug the TL072 back in. LOL

With the TL072 in its socket the circuit bursts into oscillation. You can adjust the pot in series with the two diodes to get the distortion down below 1% (I manage just under 0.7% and with REW I can see this is mostly 3rd  and other odd harmonics. I then  adjusted the second pot for 0dBu output and checked that with the output switch set to -40dB the output actually drops to within 0.1dB of -40dB. So pretty good result all round.

The track that need cutting is quite accessible before assembly. I am happy to Dremel boards before shipping them. The boards arr available to groupDIY members for the cost of postage.

I need to update the schematic to reflect the correct component idents on the PCB.

Let me know by email of you want a board or two (ianthompsonbellATgmailDOTcom)

Cheers

Ian
 
Attached is the revised schematic. I got 20 boards. I will keep two for myself. Two are already reseved. The remaining 16 are available.

Cheers

Ian
 
I have created an assembly and test document for the PCB and uploaded it to my web site. It is under the DIY tab:

https://www.customtubeconsoles.com/diy

Scroll down to the PocketOsc folder and open it.

Seven boards have now gone leaving thirteen. If you want one please email me at ianthompsonbellATgmailDOTcom

Please do not PM me because the box gets full and I don't check it that often.

Cheers

Ian
 

Latest posts

Back
Top