Broken Midiverb II: Troubleshooting Tips?

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: luketeaford, Joe., lisa, Kent

Post Reply
User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 1:28 am

OK, I built the tester and am sending a 1khz sine wave to the input, and appear to getting somewhere, maybe... Also the tester is fun, thanks!

The first two LF347N chips in the circuit (U1, U4) sound normal on all the inputs and outputs.

The op amps after the A/D/A conversion are distorted, kind of AM/ring modulated sounding.

The CD4053 inputs/outputs (that I think are audio) that sound normal are 1, 2, and 15. The other input/outputs are distorted. I think this chip is switching different signals before and after the A/D/A, I need to look at the schematic and read the circuit description again...

http://schems.com/bmampscom/alesis/ales ... %203_1.pdf

The first place I notice distortion is at the input and just before the LM311, which I think the audio signal goes through (or at least is affected by?) into the ADC, but I might be misunderstanding. The input and output pins have distortion, and so does the output of one of the transistors that feeds into the LM311 input.

*Edited to remove a lot of misunderstandings and assumptions...

Thunderbird
Learning to Wiggle
Posts: 12
Joined: Tue Dec 18, 2012 2:50 pm

Re: Broken Midiverb II: Troubleshooting Tips?

Post by Thunderbird » Wed Jul 15, 2020 6:58 am

Hi Wadesy, I ' d say there's a good chance it's that transistor, and maybe less likely one of those diodes from what you've just described.
Check the transistors and diodes with your multimeter and see.
Regards, Thunderbird.

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 12:11 pm

Thanks, I'll check those out.

Here's the description of what that part of the circuit is doing btw:
"Also note the differential amplifier (R44, R52, R53, 2 signal diodes, 2 NPN transistors) just prior to the comparator. This diff-amp
speeds up the response time of the comparator."

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 1:05 pm

Both sides of both diodes and the collector on both transistors is around +6.5v

The emitter of both transistors is -250mV, which increases as it's read

The base of both is around -25mV

Same readings with and without a signal added

The two voltage inputs at the lm311 are -11.6v and +12v

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Wed Jul 15, 2020 1:20 pm

those LM311 voltages dont seem right. are you sure its -11.6V? what are the voltages on all the pins of the LM311?
openmusiclabs.com

User avatar
EATyourGUITAR
has no life
Posts: 5405
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by EATyourGUITAR » Wed Jul 15, 2020 2:22 pm

If the diodes read 6.5v then that would be the voltage at the input of the LM311. Turn the power off, test the resistance between the diodes and the opamp. You should have two pins of the opamp that are 0.1 Ohms to the diodes. Then turn it back on. Set the meter in voltage mode. Check the voltages at the opamps, check the voltages at the diodes. There is no way you have 2A passing through a short to -11.6v and if you did, you wouldn't get 6.5v at the diodes.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 2:35 pm

1: gnd
2: (in+) +6.5v
3: (in-) +6.5v
4: (Vee) -11.6v
5: (Vcc) +12v
6: (output) +2v
7: (balance/strobe) +12v
8: (balance) +12v

Looking at the schematic and board, I think the only used pins are 1, 2, 3, 7?

There is a 4.7k resistor added on the solder side that connects pin 7 of the LM311 to the ground of an SRAM chip, 7C128-45PC. The memory is different between the 2 and 3, so I'm comparing the board of the 2 to the schematic of 3... That SRAM is socketed so I could take it out.

I'll do those other tests too

Edited: wrote pin 2 instead of 7 from LM311 to the SRAM...

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Wed Jul 15, 2020 3:04 pm

ok, that makes more sense.

by the way, pins on an IC are labeled going counter clockwise around the chip, starting in the upper left corner. so your 5-8 numberings are reversed.

when you set it to full wet, does any sound come out? does the sound vary at all with input amplitude or frequency?
openmusiclabs.com

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 3:41 pm

Haha oh man, I was literally looking at the pinout while typing and still got it wrong...

The sound coming out is pretty wild actually. There's heavy distortion (as in overdrive style) on the wet output for every preset.
The reverb presets have a loud feedback-type ringing tone.
The chorus and flanger ones have a large amount of noise (as in white noise/filtered noise).
The delay ones have audible delay happening within all the distortion.
As the input signal rises the noise level gets ducked down and rises back up on every preset.

RE: measuring resistance and voltage:
I can't say for sure I was measuring correctly... but between the diodes (from the leg after the diode) to the corresponding input pins of the LM311 resistance starts around 30 ohms and drops.

I checked the voltage again of the LM311, the diodes, and transistors and they were the same as before. (I'm measuring black probe to ground in the circuit for all these btw, I checked for ground continuity too)

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Wed Jul 15, 2020 4:14 pm

the problem is either with the diffamp/LM311, or its in the digital section. can you retake the voltage measuremens on those 2 transistors in the diffamp? the emitter being at -250mV isnt right.
openmusiclabs.com

Thunderbird
Learning to Wiggle
Posts: 12
Joined: Tue Dec 18, 2012 2:50 pm

Re: Broken Midiverb II: Troubleshooting Tips?

Post by Thunderbird » Wed Jul 15, 2020 4:41 pm

Hi Wadsey, I meant to test the transistors with the power off, sorry I wasn't clearer. Here is a link for how to test transistors with a multimeter.
You can just try testing in circuit without removing the transistors. If one of them is shorted or open circuit then you will have found the culprit!
http://www.learningaboutelectronics.com ... transistor
Regards.

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 5:00 pm

I'll check that document out, thanks! I did test the transistor resistance with power off, but wasn't super sure which points to test, so I did emitter to LM311 input.

I retested the voltage of both transistor emitters (power on...) and both measured -250 mV. I double checked I had the pinout correct too...

I also used the audio tester and switched between presets with no input signal and heard the same various kinds of oscillating/feedback/noise/distortion/etc that I did from the output at the emitter and collector of both transistors, and the LM311.

A faulty DASP-8 ASIC is kind of looking like a possibility, which I think might be the only thing in there that's not very replaceable... But hopefully that's not it!

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Wed Jul 15, 2020 8:58 pm

its going to be difficult to debug the digital section without an oscilloscope. to that end, i would suggest verifying that its not the LM311 first. one way to do this, is to desolder the 470ohm resistor (R45 on the schematic) that goes to the DAC. disconnect the DAC side, so the transistor side is still connected. then use a large value resistor (1M) to connect the free end to +12V. check the output of the LM311, is high? if so, connect the 1M resistor to -12V and check the LM311 output. is it low? if so, then thats all working fine. im assuming its all fine from your 2V reading at the output (its toggling between 0V and 5V, giving an average reading).

if its digital, you can probe all the data lines into the DAC, and see if any of them are stuck at 0V or 5V. they should all be toggling with a large amplitude analog signal applied. can you make a short recording of the output on a clean delay mode with a full scale sine input and full wet output? if its something like a stuck MSB, it might be obvious to see.
openmusiclabs.com

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Wed Jul 15, 2020 10:02 pm

Yeah, I went in thinking "as long as the problem isn't in the digital section, I can probably figure it out!"...

Is checking the 0V/5V stuck problem on the digital line possible with a soundcard-software based oscilloscope? Or is that signal at the clock frequency, 16 MHz or whatever, and not captured? I have been wanting to get an actual oscilloscope, maybe now is the time!

I attached a short clip of one of the "cleanest" delay presets with a 1kHz sine wave, full wet output. I turn off the sine wave part way through and you can hear the oscillating. A lot of the presets do that, some with a noticeable rising in amplitude like it's feeding back.
You do not have the required permissions to view the files attached to this post.

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Thu Jul 16, 2020 2:26 am

that signal has a lot of second harmonic distortion, which is common to diffamp failure. this, coupled with the -250mv reading for the emitters has me hoping that its the diffamp thats the problem. if you end up doing the comparator test described above, you can do a test of the diffamp by grounding that 470ohm resistor, and applying a very, very small signal, and listening at the diffamp outputs. the diffamp has a gain of 100 (single sided), so to keep those diodes from turning on, youll need a <5mV audio signal.
openmusiclabs.com

User avatar
EATyourGUITAR
has no life
Posts: 5405
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by EATyourGUITAR » Thu Jul 16, 2020 2:43 am

there are some ways to probe it on the cheap. you can use a chain of clock dividers to divide 10MHz down to 1KHz so that you can hear it. $3 maybe. you can use any keyboard or tuner app to check the frequency. you can't look at rise times slew etc.. it is very limited. you can buy a logic analyzer for $25. I have one here. that lets you decode i2c and SPI. or at least you can measure the frequency.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Thu Jul 16, 2020 6:24 pm

Thanks a bunch for checking that out! I'll go ahead with the comparator and diff amp tests, why not.

Anecdotally around the internet I've seen various digital problems with the midiverbs and quadraverbs fixed by replacing the filter capacitors on the +5v rail. I have the parts, but would that seem like a waste of time at this point, given these symptoms?
EATyourGUITAR wrote:
Thu Jul 16, 2020 2:43 am
there are some ways to probe it on the cheap. you can use a chain of clock dividers to divide 10MHz down to 1KHz so that you can hear it. $3 maybe. you can use any keyboard or tuner app to check the frequency. you can't look at rise times slew etc.. it is very limited. you can buy a logic analyzer for $25. I have one here. that lets you decode i2c and SPI. or at least you can measure the frequency.
That's wild, I actually wondered if dividing the clock down would work but dismissed it as dreaming.

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Thu Jul 16, 2020 8:09 pm

another easy digital tester you can make is 2 LEDs back to back (reversed) with 10k resistor in series. if one lights up, its high, if the other lights up, its low, and if both light up, its toggling.
openmusiclabs.com

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Fri Jul 17, 2020 3:49 pm

Thanks for that idea. I put together the two transistor style basic logic tester on a breadboard so I can see high, low, toggle now at least.

The digital sections are different between the 2 and 3, but I think mostly for the user interface/buttons stuff, so hopefully the schematic can guide close enough.

I'm going to research myself too, but is there a spot I should start looking for problems to work backwards/forwards from?

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Fri Jul 17, 2020 4:51 pm

have you had a chance to verify the diffamp/lm311 are good?

as for which bits to check, id probe the data lines to the DAC first, just to make sure they all toggle.
openmusiclabs.com

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Fri Jul 17, 2020 10:18 pm

I just did the diff amp/lm311 check, maybe correctly, maybe not...

I disconnected the output side of R45 that connects to pins 19-20 of the DAC, connected that to a 1M resistor and then-

To +12v: But both lights lit up on the logic tester. The output pin 7 was noisy/distorted with and without input signal.

To -12v: no lights on logic tester. No output signal on pin 7.

Diff amp test:
R45 connected to ground: input signal around 2-3mV, audible on the input op amp, couldn't hear a signal at the transistors. Checked the output pin of the LM311 and it was still noisy, no signal at the input pins.

At first I made the mistake of doing the LM311 test to R43, which is the same value but right before the diff amp, and had the same LM311 results. +12 both lights, -12 no lights

I checked the output pins of the DAC and they are toggling.

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Fri Jul 17, 2020 11:01 pm

sounds like the diffamp is bad. you could try replacing the transistors. its possible the diodes are bad there as well, but the -250mV reading for the transistors makes me suspect them. what are the part numbers on the transistors?
openmusiclabs.com

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Fri Jul 17, 2020 11:51 pm

They're 2N4401. I have some, so I could try replacing them.

User avatar
guest
Super Deluxe Wiggler
Posts: 4329
Joined: Mon Aug 19, 2013 11:49 am

Re: Broken Midiverb II: Troubleshooting Tips?

Post by guest » Sat Jul 18, 2020 1:49 am

yeah, that would be a quick try. and you could replace the diodes at the same time just to be on the safe side.
openmusiclabs.com

User avatar
wadesey
Common Wiggler
Posts: 120
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Broken Midiverb II: Troubleshooting Tips?

Post by wadesey » Sat Jul 18, 2020 2:07 am

I replaced the transistors and diodes - no change...

I remembered that the -200 mV sounded weird at the transistor, which it still is around -180-200mV, so I checked the 2 x 2.4k resistors in front of the +12 side and they are +12 in and 6.5v out. The -12v side is -12v into the 1 x 2.4k resistor -200 mV out. I think that math works out?

The problem (or one of the problems) does seem to start at those transistors, then is considerably worse at the output of the comparator. It kind of feels like there's a digital problem somewhere though?

Post Reply

Return to “Music Tech DIY”