Mark Hammer's Forty Niner sounds gated...

[dschwartz]

did you tried changing the HCF for a CD yet???

[Mark Hammer]

I'm not sure that would make any difference.

The problem seems to stem from when the frequency-booster stage is inserted, If I'm not mistaken.  Both that and the preamp stage before it are op-amps biased to Vref (or at least should be).  There is a DC blocking cap on the input to the first invertor.  Consequently, if the first op-amp provides an acceptable signal, then the second one should as well....as long as it is properly biased.

[dschwartz]

yes mark, but he tried several other cmos OD and say that they all sounded bad, so, or he made mistakes in all his builds, or, HCF chips are not as good as CD chips..wich are the things in common in all his builds..

i think it worth a shot

[Mark Hammer]

For what 4049s cost, you're right, worth a shot.  I'll have to check out some of my own builds and see whether I've used those chips at any time.

[gigimarga]

Today i will receive some CD4049UBE, i will replace all the HCFs that i used and i will tell you what was the result.
Thx a lot!

[gigimarga]

still gated...

[gez]

What pickups does your guitar have?  Are they quite hot?

It sounds as though you're either getting clipping in one of the op-amp stages (probably the second) or hard-clipping in one of the inverter stages.  Probably because your PUs are a little hot, or possibly because the op-amps you're using have limited output swing.

In the first op-amp stage, gain is set to 48.  It's easy to see that if your signal peaks at around 1V then you're going to get clipping in this stage (you'd need over 48V supply to prevent clipping...plus a chip that could withstand such a high supply).  Even if things are OK with the first stage, any gain in the second (once the booster kicks in) could quite easily tip things over the edge and produce momentary harsh clipping on peaks.  This is one of the reasons why I prefer to do all stages in CMOS fuzzers with inverters - op-amp clipping is pretty easy to achieve.  This isn't a criticism of Mark's design - far from it, it's well thought out - just a personal preference.  If you use a CMOS op-amp with rail-to-rail output swing (ICL7621 for example - mouser sell them) then it might help.

Now on to the inverters.  It's easy to forget that CMOS inverters were designed to act as switches and that they will do so if the input is dragged past the thresholds that will cause one of the devices in the inverter to shutoff.  People are often lulled into a false sense of security by the 'virtual earth' effect seen at the input of an inverter biased like an inverting op-amp: input resistor and feedback resistor.  However, what people often overlook is that once an amplifier clips negative feedback is lost.  When this happens a voltage builds up on the input of the inverter.  Buffer the input with an op-amp follower and scope it and you'll see something akin to cross-over distortion: negative earth (no voltage change) for part of the cycle and, once the output clips, a build up of voltage at the input for the peaks only.  If the input resistor is relatively small compared to the feedback resistor, then a significant voltage can build up on the input, causing the inverter to momentarily act as a switch.  The end result is a sharp flattening of the peaks that can sound like a fuzzy-honk/'blatty'. 

It's unlikely that this is happening in the first inverter as Mark has set gain for unity (with large signals at this inverter's input, gain will actually be less than unity).  However, the second stage omits an input resistor.  The virtual earth effect is nonexistent with shunt feedback bias so the input of the second inverter could quite easily be being pushed past the thresholds that cause it to act like a switch.

Solution?  Turn down the volume on your guitar.  If things clear up then you're driving the circuit too hard and need to compensate for this by reducing gain somewhere down the line.  If things sound odd/blatty even with the guitar turned down really low, then you have a biasing problem.

If you find that reducing gain cleans up things nicely but there's not enough distortion, then you'd need to add another CMOS stage (with moderate gain).

[gez]

PS  The schematic I used was found via a link from a link from the page you linked to.  It wasn't drawn up by Mark so if it's significantly different from his design then the above comments may not apply.

Hint: it always helps if you directly link to the schematic (not layouts - they're meaningless to a lot of people, myself included) of the circuit you're having a problem with.

[Edit] This schematic:

http://gaussmarkov.net/layouts/49r/49r-schem.png

[snap]

You shoulda looked at reply #2 http://www.diystompboxes.com/smfforum/index.php?topic=69078.msg553084#msg553084
(3 page PDF with layout, schematic, parts list, pcb image).


Perhaps leaky electrolytics C3 C8 C13?

[gez]

You shoulda looked at reply #2 http://www.diystompboxes.com/smfforum/index.php?topic=69078.msg553084#msg553084
(3 page PDF with layout, schematic, parts list, pcb image).

These days, if someone doesn't link to a schemo in their first post I rarely bother to look at the thread.  I don't like playing "hunt the schematic".  Life's too short...

[gigimarga]

Thx a lot gez!

I have a standard Squier Affinity (i've bought it new from a store with something like 200-300$), so hard to think that my singlecoil pickups are very hot...anyway, i rolled the volume down and the sound it's still gated...i think it's a problem with  the biasing...or with the mode of the cascading of the opamps...i have no clear ideea...maybe the "sum" of the gains of the first two stages is too big...

[gez]

I have a standard Squier Affinity (i've bought it new from a store with something like 200-300$), so hard to think that my singlecoil pickups are very hot...anyway, i rolled the volume down and the sound it's still gated...i think it's a problem with  the biasing...or with the mode of the cascading of the opamps...i have no clear ideea...maybe the "sum" of the gains of the first two stages is too big...

If this were me, I'd wire up a cap leading to a pot to ground (volume control) on a breadboard and croc-clip it to the output of the first, and then second, op-amp stage.  Feed the wiper to your guitar amp and you'll hear if there's any clipping in these stages.

Doesn't sound like that's the problem, though.  Do the voltages that you posted before still hold true?

[gigimarga]

The voltages still the same...:d

[gez]

IC1  07 - 8.35/4.51

As RG commented, presumably the two readings are for the different switch settings?  If so, then as RG said the switch wiring (or its mis-wiring) is to blame.

If pin 7 is constantly 8.35V then perhaps either R7 or R8 are the wrong value and any DC offsett at the input is being amplified.

[gigimarga]

Thx a lot gez!

I rechecked both R7 and R8 and there are OK...10K...if i bind the second leg of SW2 to VB (as in schematic) then i got a very strong motorboating...if i bind it to ground (as i did)  the motorboating disappear, but the gating appear

[gez]

Thx a lot gez!

I rechecked both R7 and R8 and there are OK...10K...if i bind the second leg of SW2 to VB (as in schematic) then i got a very strong motorboating...if i bind it to ground (as i did)  the motorboating disappear, but the gating appear

I didn't realise you were doing that.  You're amplifying DC by grounding that resistor, which is why the output of the second stage is biasing high.  To maintain correct bias, try switching the resistor to ground via a capacitor.  That way, you're only amplifying AC and leaving the bias untouched.  For a cap value, try 4u7.

[gigimarga]

Thx a lot gez, you're a genius!!!
I put SW2 to ground using a 4.7uF capacitor to the ground and now it sounds very well!
The only minor problem is a strong oscillation with all the pots maximized and the switch for the frequencies set on highs, but i hope i will solve it after i will change the value of the caps (it seems to me that the bass position is almost useless and, as i don't like very much the basses, i will try to change the caps only to have mid/highs boost).

Anyway, i built tonight an UBEsecreamer which sounds very well and i modified the 22-7 with good results, so my problems regarding 4049 circuits seems to go away

Thx a lot again!

[gez]

Thx a lot gez

Don't thank me, thank RG as he diagnosed the problem (mis-wired switch).  I just 'closed the deal'. 

[gigimarga]

Thx a lot R.G....thx a lot gez...it is possible to oscillate as i wrote above because of the value of the cap (4.7uF)?
It must to decrease or increase it? Polarized or not?

[gez]

it is possible to oscillate as i wrote above because of the value of the cap (4.7uF)?
It must to decrease or increase it? Polarized or not?

It doesn't matter if it's polarised.  If you do use a polarised cap the - pin connects to earth/ground. 

Re oscillation, you'd have to be more specific.  With the boost pot set for minimum effect (presumably this is when there's max resistance on the pot?), try turning all other controls to max.  Do you still get oscillation when switch 2 is engaged?