hello all,
I need some help figuring out my first project and its demise. i will post pictures of the board as soon as i can (website down). my transistors have obvious problems but i am not sure how to diagnose the readings and whay may be causing it. any help would be super appreciated.
thanks,
andrew
1.What does it do, not do, and sound like? The bypass works fine and sounds good but switching to the effect generates no sound but does light the led.
2.Name of the circuit = Tubescreamer from General Guitar Gadgets
3.Source of the circuit (URL of schematic or project) = http://generalguitargadgets.com/index.php?option=content&task=view&id=90&Itemid=118
4.Any modifications to the circuit? I followed a switching diagram for a grounded input with no power supply as shown here http://generalguitargadgets.com/diagrams/switch_lo_3pdt_ig_battery.gif
5.Any parts substitutions? Substitutions include using 2N3904 Transistors instead of 2SC1815 (pinout is the same), 68pF cap instead of the 51pF, and a .022uF cap instead of a .02uF cap. These are all generally accepted as equals I think. I did remove c12 as the bill of materials says this is only necessary for power supplies. Should this be jumpered?
6.Positive ground to negative ground conversion? N
(*In parentheses will be the voltages that were included in the project notes for trouble shooting purposes, followed by my voltages)
7.What is the out of circuit battery voltage? =>(8.8) 8.8
Voltage at the circuit board end of the red battery lead =8.8
Voltage at the circuit board end of the black battery lead =goes to ground
Q1
C =(8.8) 8.8
B =(3.5) 2.5
E =(3.0) 2.8
Q2
C=(8.8) 8.8
B=(3.5) 2.3
E=(3.0) 2.5
IC1 (or U1)
P1(4.4) 4.35
P2(4.4) 4.53
P3(4.4) 4.28
P4(0) 0
P5(4.4) 4.35
P6(4.4) 4.35
P7(4.4) 4.35
P8(8.8) 8.8
D1
A =4.5
K =4.3
D2
A =4.3
K =4.5
Z1 (zener)
A =
K =
[/img]
Here are some pictures. . . The wire is due to a poorly etched pad.
(http://filebox.vt.edu/users/awoodrum/pcb-bottom.jpg)
(http://filebox.vt.edu/users/awoodrum/pcb-top.jpg)
I'm not too familiar with the circuit, but it appears Q1 is not turning on because the base emitter junction is reverse biased. Notice how the stated values show +0.5 volts from the base to emitter (Vb - Ve), and you've got the opposite. I would take a look around that area and see if you've got a short or if you've used an incorrect resistor. Just my dos centavos.
Cheers,
Quote from: murdoow
Q1
C =(8.8 ) 8.8
B =(3.5) 2.5
E =(3.0) 2.8
Nice build - looks like your doing a good job....
This is just a thought... but check the data sheet on the 2n3904 you are using... I checked the datasheet for the ones I use and the collector and emitter are reversed....
It looks like you might need to turn your transistors around... but check your datasheet
If this page is correct: http://www.reprise.com/host/circuits/transistor_pinouts.asp
the 2N3904 would be positioned correctly.
Even though I put alligator clips on them, could excess heat cause the values shown? Maybe it wouldn't hurt to switch them and see?
I am currently rigging up an audio probe so we will see how that goes.
Quote from: murdoowIf this page is correct: http://www.reprise.com/host/circuits/transistor_pinouts.asp
the 2N3904 would be positioned correctly.
Even though I put alligator clips on them, could excess heat cause the values shown? Maybe it wouldn't hurt to switch them and see?
I am currently rigging up an audio probe so we will see how that goes.
Audio probe is a great tool. Just a couple of other thoughts.. The biasing is not right on the transistors as mentioned by eliktronik.. The first things I check when I have a build with issues is to make sure all of the solder holes are filed with just enough solder to fill the hole with the lead in place
You should check the voltage R9/R10 junction (R9 and R10 should be the same value) it should be roughly 1/2 the battery value. I would suspect that that voltage is off.
that is a great looking PCB (compared to mine), and i can usually find the problem after a bit of lead-bending.
the trannies look like your issue, as the base should always be about .5 or so volts above the emitter.
power supply cap? I imagine that C12 is there only to smooth out a chunky power supply. if you bridge there, all of your power will simply short! so don't! :shock: (i personally recommend leaving it there, because then if you ever add a wall jack, you're set to go... and if you have a sketchy connection to the power somewhere, the cap can run your pedal while the connection sorts itself out!)
ok, so some success and some confusion. As I was hooking it up to try the audio probe, I discovered that it does work. And sounds pretty good! (without anything to compare it to or knowing what it should sound like.) I checked the datasheets and my original 2SC1815 is ECB and the 2N3904 is EBC, (which is why it wouldn't work originally). But the values on the transistors are still backwards, C=8.8 B=2.5 E=2.85. ?
So should I call it a day or will this thing eventually crap out on me? How is it working at all? Should I swap around my B's and E's?
Also I checked the voltage between R9 and R10 and it is 4.32, or just about half.
Thanks for all the help by the way. .
You should be measuring something very close to 3.8 V on the base of the transistors (based on a quick approximate calculation). Anything between 3.5V and 4.5 V would be acceptable. The bases should be then be at about about 3 to 4 volts, and should definitely be less than the base.
There are various reasons it may be working in this condition. The most likely case is that you have the emitter and base reversed, so you're creating some kind of a forward biased diode junction looking thing that allows guitar signal through, but does not provide the high-impedance buffer input.
Definitely get your pin-out correct. The emitter of your transistor should be connected to the 10k resistor, the base will be connected to the 510k resistor, and the collector is connected to +9V
Quote from: Transmogrifox...
Definitely get your pin-out correct. The emitter of your transistor should be connected to the 10k resistor, the base will be connected to the 510k resistor, and the collector is connected to +9V
Yep, I would do the same. I build a bunch of these off of a layout related to this one.. just with tons of mods... the original always worked well for me too.
So, I would trace the emitter and base on both transistors all the way from Vr and ground - with a multimeter. The layout assumes the relationship of pins is EBC already - so that should not be an issue unless you changed that, but I suspect that ohming out the transistor's connections might expose the issue.
Also, I use 2N5088s on these - I remember using some 2n3904s once before and running into some problems like this... I can't say they had the same issue - I never really debugged it because I just slapped in some 2N5088s and it was fine. The transistors don't really color the sound in this circuit due to the emitter follower topology - so any NPN transistor that you know the pinout for should work.
As to the sound, you will know you got it right when you can run with
1) the vol and tone all the way up, and gain all the way down - you will get a big clean-ish boost
2) set the tone all the way up, volume and gain at 1/2 - you should get a decent amount of overdrive.
3) use #2 and crank the gain all the way up, you should get a low grade distortion box kind of sound with lots of volume.
hope this helps.... this is a great box with the right diodes!! Have patience with this, from my experience, any circuit new to you will have some issues to work through... Your perseverence will pay off!!:D
I swapped the base and the emitter of Q1 and now I get these values.
C=8.6
B=0.0
E=2.8
I thought I may have fried it so I put in a new one and got the same values.
The effect didn't work this time either. <Scratching head>
So I have to believe that the transistors were inserted correctly.
*Update
Just read mojotron's post and I happen to have some 2N5888s here too, as I mistakenly bought all the transistors on the bill, HA!
I will just stick those in and if that doesn't fix it, I will trace it all from the start.
A quick google search couldn't find the pinout for the 2N5888, is it EBC too?
thanks again.
I actually have 2N5088s. which were listed on the bill as "high gain" but I will put them in anyway and they are EBC.
Ok I tried the 2N5088 and now have these values:
C=8.7
B=2.7
E=3.5
So still messed up. Could something be flipping the values on both transistors Q1 and Q2?
I will try to trace it tomorrow. Ugh.
Almost all transistors are EBC, but sometimes the E and C are reversed - with respect to the face of the transistor. The older transistors from Japan will have different pinouts like BCE... but really you have to check the datasheets for all of the transistors - of just use the same ones for everything... Which is what I do - I just use Fairchild 2n5888 parts for anything that is NPN, but I use 2n3904 parts for fuzzfaces...
The 2n5088 is great... you can always add a 100-500 ohm resistor to the emitter if the gain is too much for some gain reducing degenerative feedback.
Once you have the TS working, you'll find that the opamp and diodes color the sound, well... also the caps, but it looks like you are already using the panasonic films.. those are awesome!! And, the silver-mica cap on the feedback loop is what I do too... I thought that sounded quite nice..
Yep, that sounds like the right thing to do... I would start ohming out the connections for the components conected to the bases first. - or just set the meter on a 1M ohm scale and check the resistance from the 2 bases form Vr....
Emitters should be 10K ohms to ground
Bases should be ~500K ohms to Vr
If it were me, I would also start by desoldering Q2 and trouble shooting the base and emitter of Q1 by ohming out the coneections.. but I am quick to start removing things to simplify the problem.
So does it still work?
I have to tell you, your request for input on this was perfect - impressive articulation of what you are looking at - shows that you have been around the block on a few things like this. You'll get some more input tomorrow from others who might have had the same problem, or a better idea of what might give you those readings. There is great value in the lessons learned in troubleshooting.
Update....
So I used the audio probe today. Everything seemed active and doing something except;
- the collector pins on the transistors (and that whole rail),
- pin 4 on the IC which goes to ground,
- the whole ground rail,
- +9v power lead
I also did simple resistance checks between componenets to see if it was all connected correctly and couldn't find anything wrong or broken.
I tested the transistors in circuit and they seem ok.
Per Mojotron's advice I measured resistance from the two bases to Vr (Vr is that half voltage rail right?) and got:
Q1 base to Vr= ~277k
Q2 base to Vr= ~628k
I then measured from the emitters to ground and got:
Q1 emitter to ground= kept going off the 2M chart. . . no resistance?
Q2 emitter to ground= ~107.6K Ohms ?
No idea if i am taking these mesurements correctly. . .
This made me think that the base 510k resistors were screwy but they checked out. Also the Q1 emitter seems messed up so I followed it, but it goes to the IC. So I put in another 4558D but there was no change.
I'm running out of ideas. . .
And yes, it does work, even though it tests bad. ?
One thing I just noticed, I only get readings on the transmitters if I use the ground on the input jack. I get nothing if I use the board ground.
Also I get different numbers on the pins if the effect is not engaged. duh maybe but, oh well. they are screwed up too.
C=8.6
B=5.7
E=8.3
Quote from: murdoowOne thing I just noticed, I only get readings on the transmitters if I use the ground on the input jack. I get nothing if I use the board ground.
That's part of your problem.
These are (should be) connected together, and must be at the same potential.. You either have a break in the ground trace, or a solder-bridge.
-Eric
simply connect the ground of the board to the ground of your connector and see what that does.
QuoteOne thing I just noticed, I only get readings on the transmitters if I use the ground on the input jack. I get nothing if I use the board ground.
Ok I have to take that part back. I do get the same transistor readings using the board ground. Must have been getting sleepy. :oops:
Quote from: murdoowQuoteOne thing I just noticed, I only get readings on the transmitters if I use the ground on the input jack. I get nothing if I use the board ground.
Ok I have to take that part back. I do get the same transistor readings using the board ground. Must have been getting sleepy. :oops:
Here's what I think is happening: based on the initial indications...
And you should check for 0 ohms between all of the components connected to Q1 an Q2 emitter and base(s) - looks like you might have a bad solder joint or open copper trace somewhere in the R9/R10 to Q1/Q2 base.
What this looks like, to me, 2 things that might be going on... a simple cause and a complex cause...
Simple one... Q1/Q2 emitters are not connected to ground... and you are reading voltages that are affected by the meter possibly completing the circuit... make sure the board ground is connected to the ground on the plug, and the - side of the battery - from the picture, it looks like you are missing one ground wire connection on the left hand side of the top picture... check this out and remeasure the voltages on Q1 and Q2... Also pin 3 and pin 2 should be at about the same voltage on the IC/4558.
The more complex situation is that you may be sourcing the voltage for the VR bus from pin 1 on the IC, and perhaps pin 3 as well - but pin 3 is really the base of an internal transistor (inside the 4558) so you're likely not pulling too much current from there.... (this is due to no connection/open some where on the VR bus) But, looking at the voltage at the R9/R10 junction and the higher voltage on pin 2 of the IC, with the lower voltage at pin 3 looks like you are drawing current from either pin 1,2, or 3 on the opamp, causing pin 2 on the opamp steady state/DC voltage to raise due to a current draw from pin 1....
So, you need to check all of the component conections to VR (should be 0 ohms...), including, but not as important, a check on the components that connect to them (such as the Q1/Q2 bases...).
Just a few thoughts... not being there I can't really tell for sure what's going on... but give these ideas a shot...
ok . . . so i removed Q2 to isolate the problem on Q1, as its connections looked a little dicey anyway. but am still getting the same readings on Q1:
C=8.0
B=2.5
E=3.2
I have measured resistance between components on the Vr and all traces and they check out ok. that is, if i am measuring right. All I am doing is setting the meter on the lowest Ohm setting, 200, and putting tips on either ends of the traces. it comes up in the range of 0.1-0.6. i have tried these tests for all things that should be connected; jumpers, the whole ground strip all the way to the jack, etc. . basically i have tested all the runs of traces wherever there is a long strip of copper and they all say 0. (except one point on the "Collector" rail always says -0.2?) am i measuring this correctly and (mojotron) is this what you mean by "check all of the component conections to VR (should be 0 ohms...), including, but not as important, a check on the components that connect to them (such as the Q1/Q2 bases...)."
as far as the ground goes, I have the negative terminal of the battery connected to the ring or middle post of the jack. the sleeve or inside post is connected to the board ground. the tip or post closest to the box is connected to the switch. the parts layout diagram shows another ground going to the dc jack, which i don't have. instead i have the + terminal of the battery as well as the led-resistor assembly connected to the same point on the board, as shown in the battery only diagram mentioned before.
I have also:
- checked all the parts again and measured all the resistors. all good.
- reworked any slightly dull joint.
- removed a leg of a diode and checked it out, ok.
- replaced transistor with 2n5088.
questions:
- the layout says "r6 resistor is standing on end". i took this to mean it is still connected to both pads but they are spaced so that it has to stand up.
- i am testing the transistors in this way. meter at 20 volts. mono plug inserted in input jack. effect turned on, led lights up. black tester on jack ground, red on legs of transistor.
- i am measuring resistance on the traces in the same way, with the meter set to 200 ohms.
ok ok. so i am about to put this thing in the box regardless of what my meter is saying, as it still is "working", somehow, even with the transistors showing the bias reversed. if anyone has any last minute ideas, please let me know. it has turned into a brain teaser now and its driving me nuts!
if all else fails, i bought enough parts to make two of these, and i am going to just put the other one together and see what happens. whether it works or doesn't work should answer some questions, (and i am a much better solder-er already).
thanks mojotron and everyone for all the help thus far.
Well my friend, I'm afraid I'ld have to be there to help anymore - you've done all the right things though - hate to see you get frustrated on this.
If you want you can send it to me and I'll take a look at it - PM me if you are interested - just include enough for return postage. I've been fixing/designing gear since 1980 and I have built/modded a lot of these kinds of circuits - I still have a few around here that I can use as working examples of the GGG circuit too.