0

Done building, but I only get *thumps*
Moderators: altitude, adafruit_support_bill, adafruit, hamburgers, phono

Please be positive and constructive with your questions and comments.

Done building, but I only get *thumps*

by tkahn on Thu Jun 16, 2005 4:14 am

Now I'm done building my x0xb0x. :)
I have been thorough. For example I've measured the resistance on almost all resistors before mounting them on the board to make sure I had the right value. All IC's are socket mounted (high quality sockets) and I've been as thorough as possible while mounting the rest of the parts; electrolyte caps mounted with the correct polarity etc.

I have done no tests apart from the PSU which I tested while assembling it. The values I got where similar to those in the building manual so the PSU should be OK.

This morning I plugged it in. Since I had no earphones or active speaker system I plugged in a small (passive) multimedia loudspeaker to the earphone jack. I had the x0xb0x in random mode. But all I got was small *thumps* that come with the same tempo as the tempo LED.

I opened the synth up and tried different settings for the trimpots on the main board, but that made little or no difference. When I cranked up the trimpot on the VCF part I could actually hear an increase in filter resonance but the sound was still this thumping.

I have checked the board for faulty or bad solder joints and have fixed a couple of joints, but in general it looks OK.

Any clues as to where I should be looking in order to troubleshoot my x0xb0x are most welcome! Thanks in advance!
tkahn
 
Posts: 20
Joined: Sun May 29, 2005 2:31 am
Location: Danderyd, Sweden

by Milkmansound on Thu Jun 16, 2005 9:49 am

well, go through every test like it says in the build manual and try to isolate which circuit is behaving badly. A look at the schematic will allow you to lift certain components out of circuit to isolate things even more.

However, it may be ok - hook it up to a mixer and see what the output is like. Mine did something weird for the first 2 minutes as well, then it just started to work. Be pateient - and thorough!
(((((MILKMANSOUND)))))
Milkmansound
 
Posts: 77
Joined: Wed May 25, 2005 12:33 am
Location: Brooklyn, NY

by tkahn on Thu Jun 16, 2005 12:40 pm

Here at home I have nothing but a simple multimeter (no frequency counter) for testing. The next time I will get access to stuff like an oscilloscope will be in a couple of weeks, so I guess I'll just have to wait with the troubleshooting or perhaps do some of the tests that I can do with the stuff I have.

I have left the unit running on random for a couple of minutes but nothing happens, apart from the thudding sound I mentioned.

One thing I noticed on the PSU though was that the center pin of TM6 and the closest pin of the transistor just above it was shorted. I had not clipped the component legs shorth enough and the trimpot leg was probably bent when I mounted the synth in the case. Apart from that I have found no more errors but I will take a closer look tonight.

Thanks for the input though! Right now I'll take all the help I can get. I just hope that I haven't damaged anything beyond repair...
tkahn
 
Posts: 20
Joined: Sun May 29, 2005 2:31 am
Location: Danderyd, Sweden

by Guest on Thu Jun 16, 2005 1:41 pm

if you have an amplifier of any sort you can do a lot of the tests
first listen to the mix out to make sure it isnt the headphone amp
then listen to the output of the filter as described in the manual
then listen to the output of the vco at the waveform switch
Guest
 

by Jonnay on Thu Jun 16, 2005 1:50 pm

We also had a multimeter with no frequency detection, so don't think you need a scope to test it. hell, you don't need a scope to calibrate it.

so, in trouble shooting this, I would suggest a few things:

crank up the cutoff, resonance, envmod, decay and accent pots, and see if that makes any difference. Doing this will give you a good idea of where you need to look further. Also, try playing with accent on and off. That might make a difference.

Check for dumb shit. All IC's plugged in, no little metal bits siting inside of the case, etc.

Then start with testing the VCO, as mentioned in the fab manual, use a cheap pair of headphones and probe the middle pin of the waveform switch (I think it has to be set on square, not sure about this one). Read the fab manual for details and pics. You already have a power suppy, just set it on keyboard mode, and you can send out the right voltage.

Then just work your way up the testing manual.

My guess is that something is wrong with the filter, envelope and/or accent. It sounds like your envelope is affecting the cuttoff too much, too fast, or both.
Jonnay
 
Posts: 142
Joined: Thu Jun 02, 2005 12:28 pm

by tkahn on Thu Jun 16, 2005 4:02 pm

When I set the x0xb0x in keyboard mode and crank up all the knobs and fiddle with them I hear a difference. In the headphones I can hear the filter doing the things it should to the sound, but the sound I get is just static (white noise) and it's quite faint. I have to have the volume cranked up on max.

I have checked all the IC's to make sure they are inserted the right way and find nothing wrong. Two images of my board can be found here:

http://www.maaki.com/thomas/x0xb0x_card.jpg (888 Kb JPEG)
http://www.maaki.com/thomas/x0xb0x_card_2.jpg (883 Kb JPEG)

With all the jumpers soldered it's hard to get a good picture of the whole board but this is the best I can do for the moment.

I did the PSU test again to make sure the PSU wasn't damaged and took notes this time. These were the values I got:

Large caps (across): 13.11 DCV and 14.25 DCV
IC20 (middle and left pin): 13.20 DCV
C6 (across): 5.08 DCV
C1 (across): 5.90 DCV
J4 (pin #3 and pin #2): 11.85 DCV
J4 (pin #1 and pin #2): 5.33 DCV

According to the manual all these values are within the acceptable range so the PSU seems to be working like it should.

I then proceeded with the VCO. I immediatly ran into trouble since I don't have any clip-probes. With some tape and the probes from my multimeter I attached the tip and the sleeve of the headphones to the probes. I guess it works...

The manual sais I should probe pin #1 and pin#3 of the waveform switch. I guess the tip of the headphones should be placed on the actual pin and the sleeve goes to pin #2 of J4 which is ground? The manual doesn't say where to put the sleeve probe?

Anyway I didn't hear a thing so I guess this either means I have problems with my VCO or I didn't place the probes correctly. The headphones I used where a pair of small, cheap SONY headphones. As for measuring the voltages of the VCO with a multimeter it was way to advanced for me. I didn't understand how to place the probes...

Since the VCO doesn't seem to work, I guess it's no use trying to troubleshoot the rest of the circuit.

By the way, when the x0xb0x is set to keyboard mode, shouldn't the corresponding led above each key light up when you press it? My led's don't do that(?)
tkahn
 
Posts: 20
Joined: Sun May 29, 2005 2:31 am
Location: Danderyd, Sweden

by Jonnay on Thu Jun 16, 2005 5:27 pm

When Iw as testing my VCO, I also got a bunch of static, and a very faint signal. The more I moved the headphones around, the worse the static. Your best bet is to try and keep it as still, and as steady as possible. Freaking hard. Also, the sound is going to be very faint. We grounded to R179.

I used alligator clips, and wired them together. The setup was very sketchy indeed... with one aligator clip attached to the sleeve, and the wire going from that, and being wreapped around R179...

Double check the orintation on Q24, Q25 and Q26. It looks like you might have put in 24 and 25 backwards (its hard to tell from your picture). If that is the case, the easiest way to deal with that is to clip them off (assuming you didn't solder them too close to the board), and solder-suck out the clipped leads, and remount it. That way you will minimize the risk of damaging them due to de-soldering.
Jonnay
 
Posts: 142
Joined: Thu Jun 02, 2005 12:28 pm

by Guest on Thu Jun 16, 2005 6:12 pm

your q24 and q25 are definitely backwards
Guest
 

by tkahn on Fri Jun 17, 2005 5:19 am

Hey, you are right! This is so embarrassing. :oops:

When I compare the image of my board http://www.maaki.com/thomas/x0xb0x_card_2.jpg with the one in the building manual http://www.ladyada.net/make/x0xb0x/fab/images/vco_afterd2a.jpg I can clearly see that both Q24 and Q25 are mounted the wrong way on my board. Q26 is soldered on the right way though. But since I could make a major mistake like this I will be sure to check all the transistors on the board. I hope that this haven't damaged the transistors...

I will desolder Q24 and Q25 tonight and probably do it the way you described Jonnay (thanks for the tip)!

I can't thank you enough Jonnay; you have been really helpful. :D
tkahn
 
Posts: 20
Joined: Sun May 29, 2005 2:31 am
Location: Danderyd, Sweden

by tkahn on Sat Jun 18, 2005 2:31 am

I switched Q24 and Q25 around and the x0xb0x worked like a charm - for a couple of minutes. Then something strange happened:

I was editing a pattern in pattern edit mode and i thought the tempo was too low, so I started twisting up the tempo. Then the x0xb0x froze and I could not edit anything. So I pulled the power-plug and waited for 2-3 seconds. When I started again the x0xb0x played my pattern (you could see the LED's changing) but it only played one tone; that is it was the same tone regardless of what the LED's were indicating.

So I pulled the plug again, waited a couple of seconds and put it back in. Now the LED's on the x0xb0x were lit up like a christmas tree(!) Almost all LED's where on and the machine didn't respond to any of the controls or make any sound. It has been like this ever since. I left the box overnight and tested it again this morning - still the same.

My guess is that there is a problem with a chip or the software, but I don't know... Anyway, this is a new error so I think I will start a new thread describing this new error.
tkahn
 
Posts: 20
Joined: Sun May 29, 2005 2:31 am
Location: Danderyd, Sweden

by Milkmansound on Sun Jun 19, 2005 8:14 pm

well, now you know you put some things in the wrong way - so make sure everything is on the correct way!

Also - run it and touch your finger over every transistor and IC - anything warm or too hot to touch will indicate problems. The thing should be cool as a cucumber.

Lastly, look at the schematic and check voltages.
(((((MILKMANSOUND)))))
Milkmansound
 
Posts: 77
Joined: Wed May 25, 2005 12:33 am
Location: Brooklyn, NY

Please be positive and constructive with your questions and comments.