MN3007

[scintillation]

I've built a Heladito chorus pedal and as far as I can work out, the MN3007 is clipping the signal when I strum the guitar hard.

I've got the 100k trimpot biased for around 4.5 V. The chorus sounds great, but if I play chords strumming with my guitar volume fully up, I can hear clipping on the signal (slightly on the initial decay of the strum). I've removed the MN3007 and bridged between pin 3 and 7 on the socket, and this clipping goes, so the buffer after the MN3007 seems to be fine.

I've had the supply voltage up to 13.5 V (and re-biased the trimpot) and the clipping is still there. It's not terrible clipping, but it's annoying the hell out of me.

I read somewhere about there being problems with fakes of these ICs? Do the fakes work, but just not as well, or are the unusable chips?

I don't have another MN3007 to hand, is it worth buying on to try?

I'd prefer not to have to always play with a compressor before this pedal...

Is this normal for an MN3007?

[petey twofinger]

actives ? srry... had to ask . i have a mn3007 here waiting , i use actives .

[armdnrdy]

You asked about fake 3007s.

you do get delay correct?

I don't believe that anyone would put the time and money into producing a MN3007 knock off.

Out of all of the BBD ICs produced historically....the MN3007 is still readily available in fairly large numbers, at a relatively low price.

But....they could be seconds.

[Scruffie]

Or poorly ESD handled or, you mention biasing to 4.5V... you do know half supply isn't necessarily right for the chip yeah? Sorry if that's patronising but that half bias myth gets everywhere.

[scintillation]

Using 335 dot with stock pickups.

Re biasing, yes, I've have gone through the full range for biasing, and found 4.5 V the least worst position for clipping.

I have come across DSP chips that were cheap knocks off, similar functionality, but not as tight QA on production. I think a factory had got hold of the dies and were trying to produce the chips cheaper but will less "tolerance". But the product was still labelled as a TI DSP.

And yes, get delay. If I pick notes I get perfect chorus and functionality from the pedal. It's my first time using this chip, but I find it hard to believe the MN3007 has such a poor dynamic range.

ESD is a possibility although I try my best with wrist strap etc.

Will order a new one and see what happens...

[Fender3D]

Heladito's not the best way to test MN3007 functioning...

Vgg is tied to GND and the entire circuit, expecially BIAS section, was done "in economy" to be fair...

Vgg in particular will affect BBD's distortion AND gain, but this is not the thread.. 

You have 1 easy move...
try recalculating IC1a and b gains, lowering IC1a and raising IC1b
otherwise report when distortion occours with values, this will involve those o-scope and signal generator that should have been used when trimming bias...

[Mark Hammer]

Are we talking about biasing the BBD or biasing the entire audio path?

BBDs require their own DC bias in order to pass signal.  It is NOT half the supply voltage.  As one searches for the sweet spot in biasing a BBD's input, you will pass from no signal at either extreme end of the trimpot, to a very distorted (but audible) sound, to a cleaner delay sound, the cleanest, then less clean, distorted, and nothing, as one adjusts the trimpot further.

[Fender3D]

^^

With small clone, trimming the... trimmer, means talking about biasing BBD and half audio path...

[scintillation]

I had some fresh enthusiasm for this last night. I think my conclusion is I was expecting too much from this in terms of (clean) dynamic range.

I measured the peak voltage going into the pedal. For what I would consider normal strumming, I get peaks of 1.4 V (using a scope of course...) going into the pedal (epiphone 335 dot with standard pickups). If I turn my guitar volume down to 5 (half way) I can get the signal low enough that it doesn't clip.

From my experiments I have about 1.3 V rms of clean head room (on signal going into MN3007) before I can hear the clipping (which is inline with the data sheet). I've verified this with a sig gen too. The gain preceding the MN3007 isn't doing any favours. And of course, when I tweak the first amp gain (and turn the back one up), I'm then making the SNR worse.

The problem I have is that when I'm soloing I'm not a good enough player to control my attack to the nearest 10 mV so I struggle to find a good point on the volume where I can get enough volume and not occidentally clip it with a over keen picking on a note. The clipping isn't a nice sound so unlike when you play too hard and start pushing an amp into overdrive. Although it's subtle, it sounds (to my ears) horrible.

So as I said I didn't want to do in the first post, I think compression must be used with this pedal. Are there any designs with compression built in to limit the input to the MN3007 to prevent distortion?

Is compression and better attention to the volume on the guitar the way that people get around this when questing for chorus on a clean signal? Or while I have been burying my head in the details, have I missed something obvious?

[armdnrdy]

You can try putting a limiter before the circuit in question. Try bread boarding it to see if it helps with the problem without taking away any dynamics.

[Mark Hammer]

What you are experiencing is why so many BBD-based effects incorporate compander chips.