The Rocket (again..., and problems)

[miqbal]

Hello guys,

this is my third build, but it would become the first fuzz/distortion pedal.
Looking at the schematic for the first time, and downloading the sound samples, I thought it would be an EZ build before I can move on to a more complicated distortion schemo. Well, it is not that..... EZ.

It runs for the first time but with extra bonus: lots of hiss and noisy. If I unplugged the input jack off the guitar, it starts squealing!!

I redrawn the pcb layout from JD's (GGG). I also follow Aron suggestion on socketing 3 of the caps and all the transistors as well.
The only differences are: I use F2N5089 (should be 2N5088; 450 hfe) for all transistors and 20k for volume pot (should be 100k; according to JD's bill of material). The antipop resistor is 5.6M and and all caps are the green film type except the 47u.





This is the sound of el terrible, for the curious one:):
the HISS and the NOISE
http://aronnelson.com/gallery/v/diyuser/miqbal_rocket/rocket_noise.mp3.html
SQUEALING while the jack is dangling
http://aronnelson.com/gallery/v/diyuser/miqbal_rocket/rocket_jack_dangling.mp3.html
This is the video in Flash (SWF) format
http://aronnelson.com/gallery/v/diyuser/miqbal_rocket/video.swf.html


Here are the voltage reading of the transistor (bias trimmer set on most focused sound; I measured it about 2.2k ohm):

Battery = Vcc = 8.65 Volt

Q1
C = 8.62 V
B = 4.31 V
E = 4.97 V

Q2
C = 1.317 V
B = 0.689 V
E = 0.0754 V (75.4mV)

Q3
C = 4.93 V
B = 1.28 V
E = 0.753 V



This is the analysis from a newbie:
All reading looks good except the Base  voltage for Q1. It is 4.31-4.97 = -0.66 Volt. Minus...., should be more positive then the emitter.

R.G.'s post on a thread:

.................
             The junction, at Q1 base should be at 9V*1M/(1M+0.47M) or about 6.12V. Your battery is only 8.48V under load per your measurements, so we might see 8.48* 1000/1470 = 5.76V. Further, both R1 and R3 have a 5% tolerance, and if that happened to be maximum and both in the wrong direction, you could get 5.19V. So the base voltage at Q1 is low. Not dramatically, but low.

The emitter of Q1 should be roughly 0.6V lower than the base. It's 0.7V higher. Measurement error? Confusion of Base with Emitter leads? However, you also list R4 as having 8.48V from top to bottom. If it's connected to the emitter of Q1, it should have the same voltage across it, 4.90V. So something is messed up there, either the PCB, soldering, or voltage measurement. You also list C2 as having 4.88 V at its bottom, so I'm inclined to thing you had a typo on listing the voltage across R4 and it really is 4.88 to 4.9V and there is some other error in measurement of the base. But it doesn't feel right, and I pass on just because you heard audio and because there are other more pressing issues.

....................

Well, I already double check them for short trace, pin confusion, wrong resistor value. Nope. I even have simulated a voltage divider circuit consisting only 1Meg and 470k resistor on a breadboard, the 1Meg is on negative voltage, the 470k is on Vcc. Guess what... I got (again) 4.3 Volt at the junction of voltage divider (i.e., voltage drop of the 1Meg resistor). And then I measure the voltage drop for the 470k, it only 2.3 Volt. The sum is 4.3 + 2.3 = 6.6 Volt .... Should it be 8.65 V, same with Vcc?? Where did the other '2.05 Volt' go?? Aaargh!!

This is JD's thread, also about transistor voltage reading:
http://www.diystompboxes.com/smfforum/index.php?topic=6786.0
I think it is almost the same with mine except the Q1 Base voltage.

If I was permitted to ask some questions to all the rocket guru and diy lovers here, here they are:
1. Where did the squealing and that ultra noise come from? I don't use shielded wire, because JD's don't use either (according to his rocket photo) . Is is a sign of a leaky cap??
2. I also add a 100u electrolytic on Vcc and ground rail (according to RDV's thread), adding a cap to ground at the input also,but nothing change, so any suggestion that could be taken???
3. I had try MPSA18, 2SC945, BC547, BC548, nothing change. Any more transistor that worth a try ??(actually, rocket should be less sensitive to hfe, according to Gus)

Thanks guys, any help would be highly appreciated.

PS: sorry for such a long and (most probably) confusing thread..., sorry for poor english

IQB - Jakarta, Indonesia

[Michael Allen]

hey man,

#1 thing to do, is build an audio probe. There's a diagram of a good one on RG's site, geofex.com. Trace this through your circuit and find out where the signal starts getting noisy.

I had a similar squealing problem with my RAT clone. It ended up being a faulty stomp switch that was letting a ton of crosstalk through. I would make sure all of your wiring is well soldered and as cleanly laid out as possible. Make sure all your solder joints on the board are good and shiny. I looked over the layout and it's fine.

You might try replacing the electrolytics, I think they fail more than film caps in my experience.

[frankclarke]

You don't need Q1, so take it out and connect you guitar to C2. If that works reasonably well, then something is probably wrong with the Q1 end. Which would appear likely.
Q2 doesn't look great, but one thing at a time.

[miqbal]

Hi guys,  sorry for the long reply. I was out of town.

hey man,

#1 thing to do, is build an audio probe. There's a diagram of a good one on RG's site, geofex.com. Trace this through your circuit and find out where the signal starts getting noisy.

I had a similar squealing problem with my RAT clone. It ended up being a faulty stomp switch that was letting a ton of crosstalk through. I would make sure all of your wiring is well soldered and as cleanly laid out as possible. Make sure all your solder joints on the board are good and shiny. I looked over the layout and it's fine.

You might try replacing the electrolytics, I think they fail more than film caps in my experience.

Thanks for the reply, Michael.
I do not build the audio probe, but I did the method that similar to it. I just simply touch the signal trace using my output jack cable, so I can hear how the sound is. I think I got clean sound after Q1. The sound seems noisy at the Base of Q2. How could a clean sound turn into noisy sound (or maybe more noisy..) only passing a drive pot?? I also have tried changing my electrolityc (there's only one, the 47u), but no difference.

You don't need Q1, so take it out and connect you guitar to C2. If that works reasonably well, then something is probably wrong with the Q1 end. Which would appear likely.
Q2 doesn't look great, but one thing at a time.

Frankclarke,

Great ideas pow, I connect the input not to C2 (as you've suggested), but I connect it directly to C3, after releasing the drive pot. Amazingly, the noise is highly reduced somehow, although there is still a bit noise and his. I think it could be reduced by increasing the value of bass roll off cap C4 to a higher value. So, what's up with the buffer section (Q1)?? Well, I think the sound is a little bit different with Q1 in place.

The weird thing come out. I try to change the the biased resistor to a pair of 100k and 47k (compare to 1Meg and 470k, should be the same). Voila..., now I got 5.8V at the Q1 base.
Here is the new reading for Q1:

C = 8.63 V
B = 5.8 V
E = 5.37 V

Now it is more positive than the emitter. I connect the buffer....., the squeal and the noise come out again.
Fiuh....., I almost give up here. Looks like I would be knocked out by this little circuit

Anyway, thank you so much for all great suggestions.

IQB - Jakarta

[Michael Allen]

you could have a dodgy drive pot...

[PerroGrande]

Interesting...

I think I am seeing the same thing you are experiencing when I run this circuit on my simulator.  When I give an input directly to the second stage (no follower involved), I get a solid output wave -- quite strong.  When the buffer is present, I get some noise on the output with no input and a much weaker signal. 

(On the sim, I had to reduce R7 to 10K to get Q2's BE junction to conduct, FWIW)

Must investigate further!

[PerroGrande]

Well, I hate it when this happens...

Now the sim is happy with the buffered circuit *and* the 47K value for R7.   

I'm not only interested in the problem you're experiencing, I'm now interested in this circuit.  I'm trying to wrap my (warped) mind around the biasing that is going on -- plus, I want to hear what it sounds like.  Time for the breadboard. 

[Gus]

Try a 33 ohm to 100 ohm fixed resistor between Q3 emitter and the top of trim1.  As drawn the distortion section is set to max gain because the trim pot is cap bypassed

The distortion section is Si FF type stage

Trim1 is to get different bias points and sounds.  One can use a fixed 2.2K as a good starting point. 

I like to build the buffer section with higher hfe transistors like a MPSA18.  The input resistor values are because of the 9V supply the  B to E drop makes a difference.

[PerroGrande]

Hi Gus,

Yeah -- I noticed the voltage feedback biasing sometime this afternoon while I was looking at the circuit.  That technique has always thrown my brain for a bit of a loop. 

Anyway...  I decided to take a crack at slapping one of these together, which I did this afternoon.  I built it with three 2N5089 transistors and no other appreciable changes (I added the normal leading pulldown resistor, a PS filter section (100R series, 47uF cap to ground), and a polarity prot. diode (just in case.. I've been clumsy lately).  Otherwise, I just used the default values from the schematic.  I'm so freaking lazy today -- I slapped a 2k2 biasing resistor in there, noticed I had a happy 4.6 volts on the collector, and called it good (for now).

Damn!!!  This is a NICE sounding circuit right out of the blocks!    It can go from a fairly articulate light overdrive to some serious heat just by cranking on the volume knob on my guitar.  I've gotta say -- I've been kinda looking for a circuit with this sound lately, so there is a good chance this one is getting put into an enclosure.

I'm going to mess with the tone control a bit, and probably change the curve of the gain control so that I have a bit more room to experiment with the circuit when the gain is trimmed a bit.   Not to mention listening to it a different bias points...  This is going to be fun!   

I did not experience the feedback, noise, etc, that you had, Miqbal.  I used transistors with similar Hfe to the ones you selected for your build.  I was actually pleasantly surprised by how *quiet* the circuit was, especially for having as much gain as it does.  I didn't even go nuts using the "good stuff" -- most of my resistors are just basic ol' carbon (mojo free) with the exception of the really large ones (the 1M and the 4M7 pulldown) which were metal film. 

So -- the stuff you experienced is not systemic to the circuit.  This tells me that there is something going on with your circuit board -- solder bridge, etc, etc.  Can you post a picture of your board - front and back? I didn't verify your layout, either.  Probably not a bad idea, either, as a quick check.

[Gus]

Aron did a very good job changing parts of the 3 tran to something different.

   Aron changed the gain control.  The rocket uses a volume control and sets the distortion section at max.  It has a different tuning between the buffer and distortion section and an added tone control.

Something else people can try is reducing the 10K between the stages values like 4.7K to 10K

Circuits like Aron's Rocket are more about the tuning.  One can look at and say it is a buffer to a FF type circuit with a tone section but that might mis the work, it is about high pass, low pass and gain at different part of the circuit.  Not magic parts but tuning/voicing.

IMO With Si FF types the layout and wiring is important at the higher gain setting