crate glx 65 half the volume, then slow fade to silence

[DiverDown]

{Nothing ever goes according to plan, and if it does, you've missed something - Murphy's Law Rules, okay?  :lmao: }
HTH

Dude middle name is murphy i swear. lol. :lmao:

i don't think any of the other amps i have, have a fx loop. but i'll check and definately try pairing them up through the loop on the crate. i want a scope bad. i would like the top dog high speed digital scope, but there's no way i can afford one of those, or even a crt scope right now. not much going in the piano moving business. but no use fusing. gotta get this amp working. lol yeah i know the pics are small even expanded. i guess i can try without hd next time. maybe use less space for a sd pic to keep the image size big. yes sir, a couple of diode tests and test points 9, 14, 15. a hunting i will go. 8|

well the marshall mg30fx i have doesn't have a fx loop, but it has a cd in and a emulated line out. which from the idea i'm getting from what you guys have been telling me is that those 2 jacks should be relatively the same. so i tried plugging into the crate's fx return with the marshall in the line out. no difference in volume, but the sound was distorted. i tested the 2 diodes both had 15-16v on them. tested the bridge while i was at it, and they had about 4.5v on them. i'm having trouble locating the test points though. i don't see any marked pads on the top side, and on the bottom i only see 1 or 2 empty pads.

[Roly]

I'd suggest that you avoid digital scopes initially, shop around for an analogue/CRT scope, maybe 2nd hand, more knobs the better.  You can get a new Chinese one for under $100 and people throw out their old ones when they upgrade, tap local Ham sources &c - you might get lucky and find an old Ham with a CRO they don't want who is willing to give you a flying start.  Old analogue CRO's are a biot like a stereo that feeds a tube instead of a speaker, and provided you don't poke a finger at the high voltage, pretty straightforward to repair.  I'm a bit of a CRO fanatic and have a fair collection, so if you find an old valve one that needs repair...

Finding the original circuit (PDF &c) is best, scanning is 2nd best, but if you can only photograph do not reduce the resolution, they are hard enough to read as it is (but at least we can see the general signal flow if not all the text)  Lots of light helps digital cameras.

You seem to be generally on the right track.

tested the bridge while i was at it, and they had about 4.5v on them

Not sure what you mean by that.

Just to be explicit; you found 15V across each of the zeners, yes?  If so that looks like the supply is okay.

[DiverDown]

ok, i had to go back. i'll just say i will remember zeners are the little glass looking diodes from now on. *facepalm i just meant that i tested the bridge rectifiers also. i didn't know if it would matter, but hey there it is, lol. ahem, well the zeners did not have 15v on them, more a up and down range from 0.00 to 0.065 volts?  whaaaaaa? D17, D14, D13, and D23 all did this. D28 and D29 actually that time more like 16.5v.

[Enzo]

D28, D29 are 16v zeners and ought to have 16v across them with power on in the amp.D17, D14, D13, and D23 are not zeners, and so will only show a forward drop of about half a volt at max.  In this schematic, they will probably not have any voltage across them at idle.


Any kind of diode can be glass.  Zeners could be glass or they could be plastic.   That is not a reliable way to tell them apart.

[Roly]

Oh Enzo, my eyes thank you.  <3)  :dbtu:


Okay, so it seems like the power supply rails are good.

Now you need to inject some music from your MP3/CD/Cassette player into Fx Return and see if you can get the power amp and speakers to make a really loud noise.

Post results.

{Measuring components in-circuit has many traps simply because other components are connected to them which may give confusing/misleading results.  If you really need to know then you have to lift one end, but we don't just do this all over the board because it can do more harm than good by damaging components; you really need a suspect target first, then it's worth the risk}

[DiverDown]

aww geez the diodes got me again.  i plugged the ipod into fx return and no noticeable change in volume. 

[DrGonz78]

Personally I would want to measure TP14 and TP15 for the (+/-40vDC). I would measure this once immediately after turning on the amp and then play the amp until it starts to cut out completely. Then I would test the voltage on the power side of the amp. To me it would seem something is heating up and likely could be a bad solder joint/component. I mean the power amp is lost already to some degree (half of or quarter the volume) and then after it heats up, it cuts out completely. If you are doing your tests right we can't blame the preamp for this occurrence so far.

What I think we all would like to see here all the TP9-TP13 voltages measured, recorded and reported. For another I would like to you to check all the IC(1-5) chips pin 4 and 8 for the -/+16v to see what those are measuring. Describe how you are measuring these voltages as much as what numbers you actually measured. Got to know you are doing this correct, since reading this I am a bit confused as to where we are going at this stage of troubleshooting. Take is slow and don't rush, as I don't mean to come off as rushing. Let's get down to numbers and measuring them the right way.

For example, I would test all the TP9-TP15 at the starting of the amp... Write those numbers down, and then play mp3 signal into the amp till it cuts out... Then test TP9-TP15 and lets see what we get... Are they the same or different?? My gut tells me it is something very simple and the voltages might show us better where to look for a bad solder joint or component. Personally, after doing this I would solder the entire power amp portion again. Then try the same test again.

P.S. When you run the iPod into the FX return for while does the amp eventually cut out too?? All of these questions need answers. Good luck  :tu:

P.P.S. Also here is a site that you can create a free login to get 1kHz sine wave sample that is 30 seconds long... Just create mp3 and loop that track on your iPod. That will suffice on testing the power amp circuit.
http://www.freesound.org/people/klangfabrik/sounds/28636/#

[DiverDown]

TP14, (which i'm guessing is C67 + lead to C69?) meter set to 200v DC reading is 37.2 volts. TP15, (C70 - lead to C66?) reading is 37 volts. TP9 is giving me some trouble in locating the points. i've figured out that the fx jack tip is connect to RC2 wire 5, but i cannot determine which cap is C68 due to all the glue. i'm gonna go back and try to trace that down. i'm still having trouble getting the amp to fade out like it was doing. i don't know if i should be happy or mad about that. it's been cooler than normal in the house. maybe that's not helping? the only thing i've done to the amp beside dissamble it and reassemble it 5 times, lol, is reapply solder to the power caps and touch up a grounding pad that had some weak dirty looking solder on it. gonna get back to it. just wanted to update with that while i spend a couple hours deciphering the layout of the pcb. :lmao:

oh yeah thanks enzo for the pdf. that will be great to add to my project log.  :dbtu:
and dr.gonz thanks for sharing the freesound.org link. i've actually been using that site for some time now for samples to add into music ideas. haven't ever used it as a tone generator, but now i have the perfect application for that. 8|

ok i tested TP9 as wire 5 in RC2 to C68 (finally found it) tested for continuitity, got the tone, turned on the amp 0 volts. which i have now figured out is correct. i don't on the, other hand, know how to measure AC Vpp? i tried just switching my meter over to ac, but that gave me nothing. do i need to have the sine wave running throught the amp while i'm testing for that? got the same results on TP10,12, and 13. TP11 shows being coming from the base of Q5. i looked up the pinout and datasheet for the TIP-142 and 147, and went to find the correct lead to trace and discovered there is no Q5 in the board. just an unused slot for one.

[Enzo]

OK, then you will like this too...

[DiverDown]

oh wow! dude thanks! :dbtu:

[Roly]

{Just a quicky 'coz I gotta make a trip to help push out an old school chum in his burning longboat.}

i'm still having trouble getting the amp to fade out like it was doing. i don't know if i should be happy or mad about that.

Not so much mad and annoyed.  This goes with the turf.  Faults may be intermittent, but they don't repair themselves, and this one is now hiding under the sofa.

Two points flow from this; stage gear is mission critical so it's very important to correctly identify the fault so you can be certain you have actually repaired it, not just "fixed" it.  When you do this for money these are called "bouncers", they bounce back into the workshop and your lap, and they (sometimes unfairly) shred your reputation.  Nobody cares about the other 99 that didn't, it's 100% of this case, and it really pisses clients off.

From this flows the reason we don't use the "bunderbuss method" and simply rip into a faulty item of gear randomly replacing parts in the hope.  This hardly ever (never?) works out, you muddy your own waters, can just as easily paper over a fault as cure it, and there is a very real risk of introducing fresh faults in the process and getting yourself into a real pickle (because faults don't add, they multiply difficulty - which explains why it is sometimes harder to bring up a fresh build that may have several mistakes, than find just one fault in previously working gear).

Faultfinding can be like trying to catch a skittish cat, you need to sneak up on it a bit, trying not to disturb things too much initially until you get a handle on what is going on.

I've just built a box for a well-known guitarist and it's playing up, so another part of this trip is turning up on his doorstep tomorrow armed with tools, CRO and some parts.  I want to see it in situ, as he is actually using it, because it worked perfectly on test here and his problems may be situational.  It's called "service", and when it's fixed he will tell people that I drove 200 miles to see him right, not that I'm a crap tech.  It's going to cost me more in fuel than I'll make on the job, but sometimes you just have to do these things to keep your rep.

how to measure AC Vpp? i tried just switching my meter over to ac, but that gave me nothing. do i need to have the sine wave running throught the amp while i'm testing for that?

Multimeters actually measure the average AC voltage but are calibrated to read RMS on a sinewave (i.e. it's not true for other waveforms such as clipped, square, &c).

The peak voltage is strictly;

V_RMS * root(2), (root(2)= 1.414)

...or roughly half as much again (x1.5).

The peak to peak is twice that, so strictly;

2(V_RMS * root(2))

...or roughly three times the RMS reading.  Peak-to-peak generally implies it has been measured with a CRO since that is the easiest thing to determine on the screen.

These levels are by convention with the output just fully driven (unless marked otherwise, say "typical").  In any case they are intended to give you a strong indication rather than be highly accurate.  So yes, AC signal measurements require a driving signal, however keep in mind that little DVM's are rather deaf when it comes to small signal voltages - a CRO is the ideal.

Don't worry about making mistakes, mistakes are what we learn from.

www.youtube.com/watch?v=xhQ7d3BK3KQ

[DiverDown]

i'm having some troubles with my ipod, and i keep forgetting to grab the wife's before she leaves. been looking out for a scope too. i found a $60 diy scope kit on amazon. seems dodgy from the reviews. what is a suitable scope's specs for this kind of work? do i really need to go up to or past 100MHz? a book i read suggested 100MHz. so, i've been working on a digital delay pedal project in the mean time. lol hope to have it debugged and some pics up soon.

next time i'll have some results to report.

[J M Fahey]

Experienced oldtimers such as Enzo, often suggest getting a good, working old scope, for peanuts, from places like Ham Fests, garage sales from old Techs now retired, but specially School Labs which have no more use for them.
Only precaution is to check them to work perfectly.
After all, you buy them to repair *other* stuff, not themselves

[Roly]

After all, you buy them to repair *other* stuff, not themselves

Awww ... where the fun in that?   


I assume you mean this one;

http://www.amazon.com/CanaKit-Digital-Oscilloscope-DIY-Kit/dp/B004G2KQT8

Spec;

9VDC power supply voltage
<280mA current draw 8 bit resolution
256 sample memory depth
1MHz analog bandwidth
100mV/Div-5V/Div sensitivity
1M? impedance
50Vpeak-to-peak max input voltage
DC/AC coupling
Save and display up to 6 captures to memory
Transfer screen capture to PC as a bitmap file (serial adapter not included)
Backlit LCD display

Now while this is pretty cheap it is also pretty basic.  A while back I bought an ex-ADF CRT CRO and probes for about three times this price that is very full featured (Philips PM3055) with dual trace, A delayed by B timebase, LCD setting readout, computer controlled auto-set button, and a number of other features.  I'm not suggesting you need a CRO like this, but it gives you an idea of how much you can get on the 2nd hand market for similar to double the price.

Looking at the specs above, it draws over a quarter of an amp at 9 volts, so you won't be running it on a battery, a suitable mains supply is a must, and ideally should be well regulated.

It also requires a case, and a proper probe kit like this;

http://www.amazon.com/150-MHz-Oscilloscope-Probe-Kit/dp/B0002KR9FW

1MHz bandwidth is fine for audio work but there are a couple of gotchas with digital CRO's; "8 bit" means it has a vertical resolution of 256 pixels, +/-1 pixel; it has a 256 sample depth which means its horizontal resolution is also only 256 pixels.  While these will give you a good  indication they will also appear a bit "steppy" and could be potentially misleading in diagnostic work which, apart from school labs, is mainly what you use a CRO for.

The sensitivity range on the Y axis is also not very impressive and 100mV/div wouldn't be enough for a lot of what I use a CRO for, so it would need a protected preamp.

At the other extreme 50Vp-p max input is about a tenth of a typical CRT CRO, and since I do a lot of work on valve amps would be a major liability.

It can save and restore waveforms to/from PC storage which a CRT CRO generally can't; BUT, you can't leave it connected to your PC because PC's are earthed and this will at least lead to earth loops in your test setup.  It also ties you to making ground referenced measurements when you sometimes need to float the CRO "ground" to some other reference.  And you need an "adapter" not supplied.  But how useful is this storage facility?  When I've had it available I've never actually had cause to use it, and with modern and cheap digital cameras I can take very good screen shots when I need to (which isn't often except to illustrate a point in an article).

Then there are the customer reviews;

Reckless and negligent business practice
"The seller is well aware of design error"

CanaKit DSO 062 Oscilloscope
"I would not recommend this little oscilloscope ... it does NOT work ... There is NO customer service ... I would not recommend CanaKits"

Good kit if you're advanced
"Persons with no experience in electronics should not get a kit. ... your directions for building it were not that clear."

Miss Led
"the kit did not include a power supply, case, probe or instructions."

You need to be in the right frame of mind
"This kit is definitely not for everyone. Do not even attempt it unless you are well experienced putting electronic kits together."

So all in all I'd consider this a gadget, toy, or demonstrator, not a serious oscilloscope.

I have reviewed a number of freeware oscilloscope programs for Windows that employ the sound card input.  These have most of the limitations of this kit, plus they are AC only.  They particularly suffer from the common grounding problem, only one is actually voltage calibratable, and unless you build a protective preamp/attenuator to go in front of your sound card you place your computer at risk.  I get around this by using old junkers like 386's to P-1's running W95 to XP as my workshop computer, of which I have a ready supply free, so if I blow one up by doing something stupid then it's no big deal - but they only have a limited role anyway (e.g. spectrum analyser, data recorder) since I have a range of analogue test gear available which are safe to 500 or more volts with ground isolated signal commons.

http://www.zen22142.zen.co.uk/Prac/winscope.htm

Overall I agree with JM, shop around 2nd hand sources - there are good CRT CRO's going for cheap prices thanks to the infatuation with digital CRO's - take advantage of it. {Jaycar here in Australia are currently clearing full-featured $600 CRT CRO's for half price.}

[Enzo]

My thoughts:

ANY scope will work fine for audio in terms of specs.  They do not make a scope with too low a bandwidth or freq response.  When was the last time you saw a scope with LESS than 1MHz bandwidth?  I recall some fine old Tek scopes with qa 500kHz response.  1/2 of a megahertz.  Even that is way more bandwidth than you need.

Are some scopes more desirable than others?  Sure.  Are some guitars more desirable than others? Yes.  But you don;t have to buy a $4000 guitar to play, a useful old beater will work just fine and set you back $100.  If a scope puts a trace on the screen and actually responds to the input, it will be useful.   My point is that it is MUCH more important that a scope be working than it is what its specs are.

I much prefer triggered sweep.  It will stabilize the image better.  But older scopes, especially entry level ones, could be "recurrent sweep."  That means the horizontal sweep is not linked to the signal, it just sits there sweeping at whatever speed you set.   SO your sine wave might slowly creep along the screen instead of sitting there rock steady.  well even if that picture slowly sweeps along - like the heart monitor scope in a hostpital - you STILL can see the ripple on a power supply or the clipping on an audio signal.  And that is what the scope is FOR.   Most scopes are triggered these days, but don;t pass up a cheap or free one because it lacks triggering.

Like a cheap guitar, a cheap scope will teach you how scopes work and what they do and what they are good for.   Someday you may be ready for a fancier scope amd by then you wilol know what you are looking for.


Just my opinion, but I don;t think digital scopes are for beginners, and frankly maybe not the best choice for audio anyway.  We associate "digital" with new and up to date and high tech, but digital scopes have their own characteristics, and if you do not understand those, the results may be confusing or less than helpful.

Now if you are doing serious work, and especially work in the digital comain, then maybe to consider a more powerful scope, and/or a digital one.  But for straight amplifiers, not needed.

I agree with JM agreeing with me, look at old used scopes for cheap.