I've built some kits, and can use a soldering iron, but know almost nothing about trying to troubleshoot a battery-powered electric guitar effect. Its an Electro Harmonix Electric Mistress Flanger, and it consists of about two dozen components. I can find where the power source connects to the board. I hit the first IC and I don't get a signal, but am I doing it right? How do I test it?
I built a signal probe. I plug a CD player into the "guitar in", and touch the probe to the PCB trace (which connects to an audio amplifier), and the music comes out of the amp if I'm following the trace correctly. Will that help me find the bad component?
I should be able to trace power, no?
How do you follow a trace on a PCB when it hits an IC or cap or resistor? Do you have to disconnect it or ?
Whats a good reference for troubleshooting PCB boards and circuits? Either on the web or in a book?
I expect you'll get lots of suggestions from other users of this NG, some helpful perhaps. My contribution is:
(a) get the schematic of the model you are working on from the web (e.g.
formatting link
and, if you need to, data sheet(s) for the semiconductors used (e.g. for the SAD1024 in this case)
(b) figure out how the circuit works from the schematic and the data sheet(s), and therefore what kinds of signal or other voltage should be found at different points (i.e. junctions) in the circuit
(c) correlate components in the schematic with components on the PCB and then you'll know what you should find by probing the PCB - you'll also answer your own first two questions
A bright lamp is useful for following tracks on a single-sided PCB; perhaps even a double-sided one if it doesn't have expanses of ground plane.
you have to know what the ic does. It has inputs and outputs... to follow a passive component you just "go through" the component. (same thing sorta happens inside the ic but its extremely complex and you don't know the "connections" so you have to know the macro function)
First look for the obivous. Any burned components or burnt smell? Any broken wires or traces or loose components?
Look at what is happen? No power? No sound? Distorted sound, etc...
Look at the input, output and power stages. This are the ons likely to go first. Components such as resistors and capacitors rarely go back so they can be ignored at first(although if they do go bad and it's not obvious it can be very frustrating figure it out). Digital logic also rarely goes bad.
Thing's like diodes, transistors, regulators, op amps, etc.. are the first bet.
The main thing is to test power.. because it is the easiest and obviously if it's not getting power your not going to get anywhere and this is really the place that gets screwed the most(people trying to power it with different voltage than it's rated for).
In fact I'd test the power pin of all the IC's starting with the power stage and at the same time test the diodes(which you can try to do in circuit but doesn't always work but might lead to a quick answer).
What you'll probably find is that either the power stage or discrete transistors are bad but it could be something as simple as the input or output comming loose or a connection breaking do to wear.
You say you hit the 1st IC and yhou dont get a signal. You need to know what that IC is and what pins to test to see if your audio should be travelling down that path. DId you get the schematic that was linked in an earlier reply? Also, what's wrong with the flanger? Knowing what the symptoms are can help us to point you in the right direction. It's hard to suggest a plan of action for testing when we dont know what the problem is.
It really would help to see the schematic, especially since all kinds of things get misnamed.
Since by definition a flanger, if this really is a flanger, the audio is split into two paths, one going through a delay line and the other going directly to the output, where it's combined with the output of the delay line.
Hence, if there's no audio at the output, then it's likely something really simple. Either an early buffer stage, or a stage at the output that combines the two signal paths. Otherwise, there would be audio at the output, just no flanging effect if there was something wrong with the delay line.
Which of course reinforces the need for understanding before troubleshooting. If you can look at a schematic and see "oh, that's just buffering the input and that resistor brings some of its signal to the output, then it's less daunting than seeing this big expensive IC and assuming that must be where the problem lies.
As some one else here had provided a schematic (nice of them :)), I took the liberty to examine it.
The device, being nothing more than what I call a reverb unit more or less, could be a little beyond your current technical skill level to properly debug the problem with out future possible damage using the methods you have out lined. In case you didn't get the link to the schematic provided by another poster here.
formatting link
List of things I would check for.
#1.
Pin's 12 and 14 of the SAD 1024A chip need to have a square wave or near to it pulse train signal to get the serial chip moving is data down the registers. Not sure with the present equipment you have at hand how you're going to determine this ? ;/ Maybe if you had a logic probe it could help using the pulse LED option to determine if its running.
And the rest of the idea's you can ask for if you think #1 isn't beyond your skill set.
P.S. If you plan on doing this kind of work, you really should think about investing in a multimeter with Freqenency,Capacitance, Resistance, voltage, amps etc... You don't need a high end unit for what you're doing.
That's assuming the schematic the original poster is using is the same. We don't yet know that, it's possible he has a schematic that is labelled as the device, when it's not. Or, there may be multiple schematics of the same commercial device, but which come from different sources so they are not all quite alike. Ie, if a bunch of people traced out the circuit, and the put it on the web, there's no guarantee that they all match since someone might have made an error in the tracing.
Well no. It would be called a reverb if it was that. It's all in how you combine that delay with the original signal, so the waveform actually changes. You are talking about a small delay, so the delayed signal summed with the original will cause a new waveform. Note that the sound effect requires the delay to be varied continuously.
I'm not sure we've really heard from the original poster about what isn't happening. I don't recall it being clear from the original anything beyond "it's not working", which is ambiguous since it could mean the whole thing is dead, or it could mean he doesn't get the flanging effect.
The whooshing sound he's expecting to hear won't be there unless whatever varies the element that controls the delay is functioning. Like I said, because of that mixing at the output, there should be an output signal whether the delay is working or not.
SO if there's no sound at the output at all, he either doesn't have power to the devices, or something is wrong with the input buffers (assuming the schematic he is following is about what many websearch hits show).
If he has output sound, but none of the effect he's expecting, then it's the delay that needs attention, be it whatever the making the delay (the sad1024 if it's the common schematic), the clocking of that delay IC, and maybe whatever is varying that clocking of that delay IC.
Since the delay is so small, he won't be able to tell if it's working by waiting to hear a sound at the input come out a tad later. He'll only be able to tell if there's a different sound at the output.
So if there's sound at the output, he ought to disconnect the delay from that output, put in a temporary switch or something so it's easy to compare, and then see if switching out the delay causes a change in the sound. That's not definitive, but if he hears something, that helps to narrow down where the problem lies. If there's no change, then likely there's no delay in effect, the reason to be determined.
Figure out the frequency of the clock to the delay line (once again assuming it is that schematic in circulation) and if it's in audio range, one can carefully attach an audio amplifier through a coupling capacitor to the clock and see if they hear something.
Very nice of them, indeed! Thank you Christofire!!
Unfortunately, the schematic referenced on DIYSTOMPBOXES was the
*Deluxe* EH Electric Mistress. Mine is just the Electric Mistress, 2
- 9v version circa 1979. I bought it brand new, but it sat for a long time and when I pulled it out to try it, I get no sound out of it.
To be more precise, I plug guitar into input, and plug output to amp. With the footswitch off, I get a clean signal, footswitch on (FX engaged), I get nothing - no sound at all. Originally, opening up the FX I found what looked like the positve power wire from the batteries had broken off from the PCB. I looked at it with a mag glass and found where I thought it should go and resoldered it. Didn't help. Maybe its in the wrong place?
I scoured the net and found a couple of the schematics (that *used* to be on DIYSTOMPBOXES.com, but have since disappeared), but I guess I can't attach them to my posts, right? Can I upload them to someone who is hosting so I can have a real link to show what I mean?
I'd also like to show the front / back of my PCB to show what I've done, which may not be correct. I don't know how to relate a schematic to the actual circuit which is probably essential, right?
Maybe these schematics aren't right? I'm not finding the components on the PCB where I think they should be according to the schematic.
Do I understand (thanks to Jon Slaughter) that this whole problem might be due to a bad diode, or transistor? I'm going to have to round up data sheets on the IC's involved and see if I'm getting power out of them.
I have purchased two $35+ multimeters (probably chinese) that claim to be able to test transistors and capacitors, but I'll have to remove them to test. Is this where I should start? Testing each of the components values? Or is there a way to test them in-circuit? Or should I pull the diodes and test them first?
Thanks to you all for weighing in on this. Maybe this effect *will* work again someday!
Very nice of them, indeed! Thank you Christofire!!
Unfortunately, the schematic referenced on DIYSTOMPBOXES was the
*Deluxe* EH Electric Mistress. Mine is just the Electric Mistress, 2
- 9v version circa 1979. I bought it brand new, but it sat for a long time and when I pulled it out to try it, I get no sound out of it.
To be more precise, I plug guitar into input, and plug output to amp. With the footswitch off, I get a clean signal, footswitch on (FX engaged), I get nothing - no sound at all. Originally, opening up the FX I found what looked like the positve power wire from the batteries had broken off from the PCB. I looked at it with a mag glass and found where I thought it should go and resoldered it. Didn't help. Maybe its in the wrong place?
I scoured the net and found a couple of the schematics (that *used* to be on DIYSTOMPBOXES.com, but have since disappeared), but I guess I can't attach them to my posts, right? Can I upload them to someone who is hosting so I can have a real link to show what I mean?
I'd also like to show the front / back of my PCB to show what I've done, which may not be correct. I don't know how to relate a schematic to the actual circuit which is probably essential, right?
Maybe these schematics aren't right? I'm not finding the components on the PCB where I think they should be according to the schematic.
Do I understand (thanks to Jon Slaughter) that this whole problem might be due to a bad diode, or transistor? I'm going to have to round up data sheets on the IC's involved and see if I'm getting power out of them.
I have purchased two $35+ multimeters (probably chinese) that claim to be able to test transistors and capacitors, but I'll have to remove them to test. Is this where I should start? Testing each of the components values? Or is there a way to test them in-circuit? Or should I pull the diodes and test them first?
Thanks to you all for weighing in on this. Maybe this effect *will* work again someday!
You could post them to alt.binaries.schematics.electronic where schematics and photos are more welcome, with a note relating each post to this discussion in this NG. You might pick up a bit of additional interest there as well.
Have you tried to find the schematic for the model you have using Google and otherwise 'scouring' the net? ... I'd recommend it if you haven't yet.
They *used* to be on DIYSTOMPBOXES.com. They have dead links on that site now. However, I downloaded them sometime ago when the links were still good.
Thanks much for the urls to post the photos. I checked with my ISP and they have a picture share area. I might give that a try, but if it doesn't work, I'll go with one of the sites you provided. I don't use a newsreader so I can't really upload schematics / photos to alt.binaries.schematic.electronic, though it would serve the community better. Once I get setup to do that I will because I'm sure there are others looking for that schematic.
Using some sites that JeffM has suggested, I was able to get on one that was relatively painless. Please feel free to take a look at Stellan's schematic of the EH Electric Mistress (2x9V battery version). I'm not sure of the accuracy of the schematic. I still have to shoot the component side of the flanger and then I'll post both the solder side and component side (if that's a help).
As you had mentioned, I'm getting no audio on the output side when the flanger's engaged. If you can tell me what components are involved in the buffering stage, and the output stage that would be a big help. I can try troubleshooting those portions.
My guess is that the oscillator isn't oscillating. That will prevent the sad1024 from working.
Do you have any test equipment at all? An AC voltmeter will tell you if it is oscillating. Measure the AC voltage between pin 10 and 11. If it is 0, then you have no oscillation. Comnpare the voltage of the various output pins of the comparator (LM339) to ground (pin 10 on the
4013).
If there isn't any oscillation on the comparator, then that is the problem. Check the circuit around the comparator for cold solder joints, etc.
If there IS oscillation, see if it is on pin 10 and 14 of the sad1024. If it is, you sad1024 is bad, or not connected to the input for some reason. See if the signal is getting through to the SAD. The analog signal enters at pin 15.
Thanks Bob - I'll check it out. But as Michael Black mentioned in his post - if the SAD 1024 was bad, wouldn't I get some kind of audio thru the output instead of none at all? The signal would be clean, with no FX but it would be something, right?
As Monsen said, it has to get thru the 4558/4558 stages first to get to the 13k. If one or both of those opamps are shot, you get zilch. . . ...and you sure do blockquote a lot more than is necessary.
If the SAD 1024 is dead nothing will be at the output, other than a DC voltage. If there is co clock, the signal can ot be transferred from the input to the output. That part of the circuit is digital. That is why they chip is a called a 'Bucket Brigade'. The sample is transferred to the next analog memory cell, then refilled, through all 1024 cells.
--
http://improve-usenet.org/index.html
aioe.org, Goggle Groups, and Web TV users must request to be white
listed, or I will not see your messages.
If you have broadband, your ISP may have a NNTP news server included in
your account: http://www.usenettools.net/ISP.htm
There are two kinds of people on this earth:
The crazy, and the insane.
The first sign of insanity is denying that you\'re crazy.
Actually, he is right. I misread the schematic... the point of a flanger is to add the 'current' signal to a delayed signal. So, even if the delay chip is hosed, he should hear the input, without flanging effect, if the SAD is dead or if the oscillator isn't oscillating.
So, it is probably the input stage that is AFU. The OP should check the input to the SAD, and work backwards towards the input to the box to see where the signal is failing.
The OP could also check whether there is signal on the output of the opamp (the 4558 in the schematic). That is pin 1 or 7 of the 4558 if you are using one of those. The power pins are 4 and 8. 4 should be at ground, and 8 should be at Vcc, whatever that is, probably 9V.
This is a dumb question but looking at the data sheet and the schematic for the SAD 1024, does pin 2 (IN A) feed one set of 512 buckets, and pin 15 (IN B) feed the other set? If so, should I get signal on either of those two pins?
AFU=I think I know what that means? OP=?
When you say signal, does that imply using an oscilliscope (which I don't have), or are you talking about actually "hearing" a guitar through that pin with some device (like an audio probe)?
Yes. The 4558 is >As Monsen said, it has to get thru the 4558/4558 stages first
OK - I see that. The leg with the 13k ohm resistor is the bypass, correct? That makes sense now. . . "...and you sure do blockquote a lot more than is necessary. "
Probably a carryover from my job. I have a habit of trying to be as verbose as possible.
Thanks again gentlemen for mapping my next moves. Rob
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.