Building a Black Eye : tubes daughter boards mistake, is it really bad ?

[pbrommer]

Honest question: Did the drop in volume follow when you switched tubes? If so, then bad tube. If it did, then bad tube.

Things to check would be the voltages on both sides of R11 and R12 (the main resistors to the tubes). The two voltages closest to the SushiBox label and the 1n and 470p capacitor will tell us the voltage from the banded side of the UF4007. It will help us diagnose where the issues might be.
Voltage of R11 and R12 nearest the tube socket will tell us if the 100k are okay and the correct voltages are getting to the second tube's plates correctly. Double check that they - R11 and R12 - are 100k!

Also, can you measure the voltage of the two sides of the UF4007 diode? This could have had an issue because the tubes were the wrong way and it might have pulled too much current and shorted the diode. I've had that happen before!

Another idea would be to pull the NE555 and replace it and then recheck voltages. If those still have the same problem, we might need to pull V2 and look for issues.

I have a couple more ideas to check, but really voltages help. It's something I've learned from numerous screw ups and building a few amplifiers here and there. Yes, tube amplifiers are scary and there are high voltages. Best practice -- put the negative lead in the screw hole of the enclosure and hold the red probe in your dominant hand. Put your other hand behind your back when checking voltages. This will prevent voltage from going across your arms and across your heart. After working with tube amps and getting zapped by 350VDC+, I'm cautious. But it can be solved! You'll get it solved!

 

[eh là bas ma]

Did the drop in volume follow when you switched tubes?

Thanks for your reply and support !

Switching tubes didn't make any noticeable change. The volume was still as loud as before and as low as before on the three different locations mentioned by Vigilante. So i guess the tubes are good at least.

can you measure the voltage of the two sides of the UF4007 diode? This could have had an issue because the tubes were the wrong way and it might have pulled too much current and shorted the diode.

I don't think so, I didn't connect the power supply to the circuit when the tube daughterboards were backward. I didn't even solder the tube sockets.

When i noticed my mistake, i posted and waited for a reply before doing anything else. First time i plugged the power, their orientation was all right, if i recall correctly... it was 2 or 3 weeks ago. Could have been a short in this aera though...
To make sure, I will unsolder the UF4007 diode and check it with my DMM on diode mode, and report.

Then i will put back the unsoldered resistors, proceed with the voltage readings as suggested, one hand behind my back, and report them.

Maybe this thing just need a little push... It does feel like it's almost working...

 

[pbrommer]

Okay. Don't pull any resistors or diodes yet! Measure them before you pull them, because then we can diagnose where the problem is. And I'll admit, I read through what you did with the tubes and I assumed you had plugged it in with the daughterboards incorrect.

If voltage is good on both sides of the diode, then we're awesome (I'm assuming so, because the first tube sounds fine). But please check the voltage on both sides of the 100k resistors on tube 2. That's where we're having an issue. There's either a wrong value, shorted component, or a bad component somewhere. We've isolated WHERE the problem is. Now we need to isolate WHAT the problem is. You're right: you're super close to having it working. If you're like me, it's going to be the dumbest possible thing that is the problem.

 

[vigilante398]

In all the tube builds I've done (there have been a good handful) I've never seen the rectifier fried, the MOSFET always goes first if there's a heavy enough short on the line. In order for the SMPS to generate enough current to fry the rectifier with a heavy load on it the MOSFET first has to switch hard enough to get there, and that's a big ask.

Also note that when measuring the anode side of the rectifier you'll be getting AC voltage at around 40kHz and some meters can't read that very well. If you're getting 180VDC on the cathode side then your rectifier is likely fine.

My best bet is that the traces may have been damaged when the daughter boards were disassembled and reassembled. Next time I send a shipment to Klaus I'll see if he'd be interested in stocking spare daughter boards so he can supply those in case people need them.

 

[eh là bas ma]

I've put back R21 and R18 in the circuit, after making sure again that they have their correct values.

Circuit powered, switched off. Black probe on input jack's ground. ("screw hole" you mean where we close the back of the enclosure ? I have a crocodile clip on my probe, so it's too narrow. Is the enclosure better than the jack's ground terminal ?)

At UF4007 negative side i read 160V
On the positive side (near the 3PDT) i get no reading, nothing or just random values briefly and back to nothing. The meter keeps searching, it doesn't read 0L or something, just blank but kind of busy.

R11 near the logo i read 152V. Near the daughterboard i read 124V.
R12 near the logo i have 152V. Near the daughterboard : 112V

I will unsolder the UF4007 diode, make sure it's ok...

Edit : i switched the meter to AC voltage to probe UF4007 + side. I read 0.050V.
I checked the diode without unsoldering it, i read 0.600V, looks fine i guess.
Also checked R11 and R12, still soldered to the circuit, they both read 99.5k.

 

[pbrommer]

In all the tube builds I've done (there have been a good handful) I've never seen the rectifier fried, the MOSFET always goes first if there's a heavy enough short on the line. In order for the SMPS to generate enough current to fry the rectifier with a heavy load on it the MOSFET first has to switch hard enough to get there, and that's a big ask.

Also note that when measuring the anode side of the rectifier you'll be getting AC voltage at around 40kHz and some meters can't read that very well. If you're getting 180VDC on the cathode side then your rectifier is likely fine.

My best bet is that the traces may have been damaged when the daughter boards were disassembled and reassembled. Next time I send a shipment to Klaus I'll see if he'd be interested in stocking spare daughter boards so he can supply those in case people need them.

I'm not well versed in SMPS, so didn't know there was AC on the 555 side of the UF4007. The more you know.

 

[music6000]

Can you take a voltage reading on the 555 pins as numbered?

 

[eh là bas ma]

NE555

1= 0V
2= 4.29V
3= 4.56V
4= 8.91V
5= 5.93V
6= 4.30V
7= 4.30V
8= 8.91V

Looks good too ?

 

[eh là bas ma]

My best bet is that the traces may have been damaged when the daughter boards were disassembled and reassembled.

That would be my guess too, but i was extra cautious whith the pcbs. I have a soldering-residue cleaner in spray, with a brush on the mouth, and i cleaned everything so the boards would look clean, even after the dismounting.

Shouldn't i notice something with a visual inspection if a trace was damaged ?

I tried earlier to do some continuity tests between the connectors pads on the main pcb and the tubes sockets pads. I didn't know exactly the connections, but all the sockets pads were ringing with some of the connectors pads.

Is there a way to make sure that a trace is damaged in this aera ?

Sockets have 9 pads, connectors have 10, so i guess it should be quite simple to find the bad trace if there is one. I guess the 10th connector pad is for the led.

 

[pbrommer]

I've put back R21 and R18 in the circuit, after making sure again that they have their correct values.

At UF4007 negative side i read 160V
On the positive side (near the 3PDT) i get no reading, nothing or just random values briefly and back to nothing. The meter keeps searching, it doesn't read 0L or something, just blank but kind of busy.

R11 near the logo i read 152V. Near the daughterboard i read 124V.
R12 near the logo i have 152V. Near the daughterboard : 112V

I will unsolder the UF4007 diode, make sure it's ok...

Okay, that's a great start. According to Mr. Sushi , we have AC going to DC over the UF4007.

Voltage drops across the resistors look okay. So now to figure out why there's not 250vdc. Looks like, to me, the issue is in the areas coming off the NE555 - the inductor, those resistors, and the MOSFET.

Also, it looks like the low volume is coming off of the tube, so in the volume know/tone knob area. So, we've isolated it, now to get the solution!

 

[eh là bas ma]

I did some continuity tests on the right daughter board, the one near the Sushi logo, output jack side :


So i have 1 socket pad and 1 connector pad that don't ring with anything, and one connector pad that rings with two socket pad ( in dark blue).

On the other daughter board, i did a quick check, all socket pads are ringing with some connector pad...

Should i do something here ? connect these two uncolored pads with a jumper ?

On the instructions layouts, top and bottom, i can't see any trace going to this connector pad ?

 

[pbrommer]

I did some continuity tests on the right daughter board, the one near the Sushi logo, output jack side :
View attachment 63200

So i have 1 socket pad and 1 connector pad that don't ring with anything, and one connector pad that rings with two socket pad ( in dark blue).

On the other daughter board, i did a quick check, all socket pads are ringing with some connector pad...

Should i do something here ? connect these two uncolored pads with a jumper ?

On the instructions layouts, top and bottom, i can't see any trace going to this connector pad ?

Edit : If someone who built a Black Eye read this : Can you please confirm that these two pads, without colored circle, are connected together ?

Maybe even comfirm all of these continuity tests on the right daughterboard ? Please it should only take a minute ? For me it's been 3 weeks ?

If I read that pin out right, that should be pin 7 of the second tube, exactly where the drop of volume is happening. Check for continuity from the board to the pins of the socket on the other side and even the tube pin.
Those should all connect back to the 220k next to the 330k. Maybe check if there is continuity there - tube to the 220k? If not, we might have found the problem spot. There might need to be a jumper there.

I'm giving this my best shot based on my knowledge of lots of stuff but not much expertise. It's almost solved: we have the location but not the problrm spot.

 

[eh là bas ma]

R17 220k on the 3PDT side, is ringing with the pink socket pad and the pink connector pad. The ringing is steady with the connector pad, not so much with the socket pad, i have to insist a bit to hear the ringing and it's not very steady.

Edit : still no idea about the two uncircled pads from the picture reply #51. I'd like to risk everything, and connect them together but christmas is coming soon...

 

[vigilante398]

Were those continuity readings taken with a tube installed? There should be no connection to pin 9 of the tube, and pins 4 and 5 should be separate.

The uncircled pads should be connected, there's a trace with a via, it jumps layers.

 

[eh là bas ma]

Were those continuity readings taken with a tube installed?

Yes they were.

There should be no connection to pin 9 of the tube, and pins 4 and 5 should be separate.

So one socket pad shouldn't be connected to the connectors, and i am guessing pin 4 and 5 are the dark blue ones ?

Just to be sure that we have the same designations, is this the right pinout diagram ?

The uncircled pads should be connected, there's a trace with a via, it jumps layers.

You mean connected together ? Sorry if i'm a bit slow, first high voltage build....

 

[pbrommer]

Yes they were.

So one socket pad shouldn't be connected to the connectors, and i am guessing pin 4 and 5 are the dark blue ones ?

Just to be sure that we have the same designations, is this the right pinout diagram ?
View attachment 63230

You mean connected together ? Sorry if i'm a bit slow, first high voltage build....

That is the correct pinout. 4 and 5 are NOT connected when you power the tube with 12.6vdc/vac. They ARE connected if you use 6.3 vdc/vac. In this pedal, it's using 12vdc, so no connection to pin 9 and 4 and 5 should not beep if you check continuity between them.

I think Sushi said connect the two uncircled pins. If I'm right, that's pin 7, which is where you're having volume drop off issues.

 

[eh là bas ma]

Klaus from musikding answered me today and agreed to send me a fresh pcb set for 24 euros.

I guess it's better to wait for it, and make the cleanest, safest build i can do.

In the meantime, i'll try to fix the socket's connections according to your instructions, just in case.

Thanks for your help and support !

 

[pbrommer]

Klaus from musikding answered me today and agreed to send me a fresh pcb set for 24 euros.

I guess it's better to wait for it, and make the cleanest, safest build i can do.

In the meantime, i'll try to fix the socket's connections according to your instructions, just in case.

Thanks for your help and support !

Cool. Perhaps after you rebuild it and get it running, we might be able to find the issue with the first one. Then, Stereo BE!

 

[Dysvalence]

With the caveat that I might just be talking out of my ass on a legitimately dangerous topic, I'm wondering how the voltage dropped to 170V in the first place, and how R20/R13 is more distorted than R17/R22, when V2B on the schem is the cold clipper- afaik the main distortion stage.

Assuming the latter is really just from C6 rolling off the highs and making things sound smoother, I'd then suspect the feedback part of the SMPS- if Q1 was already replaced maybe the issue is with R2/R5/R6. I would wait for a second opinion before doing anything though.

 

[eh là bas ma]

I added a jumper between the two uncircled pads : the volume and the signal strength has been restored ! Hurrah !

On the musical side, now the effect sounds fully working, and ready to be put to good use.

However, i can't find the short between pin 4 and 5 (dark blue). I moved the tip of my iron all around the dark blue connector pad on the daughterboard and on the main pcb, but it's still ringing. Visual inspection didn't help neither, i can't see anything suspicious.

Even if i remove the tube from the socket, these two socket pads are still ringing.

I checked the other tube daughterboard at pin 4 and 5 and i also have continuity there.

If pin 4 and 5 shouldn't be connected together, which one shouldn't be connected to the dark blue connector pad ? And to which connector pad should it be connected instead ?

Voltage at UF4007 negative side is still at 162V instead of 240V.

Just to be clear, these continuity tests have been performed on the soldering side, the tube is located on the other side of the pcb. In the left picture with colored circles, we are looking at the soldering side of the daughterboard. The tube socket is set on the other side, with a white circle instead of a Sushi logo :

There should be no connection to pin 9 of the tube

Pin 9 is the purple socket pad, above the "i" of the Sushi logo ? On my build it's ringing with the purple connector pad, i checked the other daughterboard and it also rings with the same corresponding pad. I can also see a trace on the pcb between these two points.

If i am being confused with the daughterboard's orientation, and pin 9 is the light blue socket pad (above the "S" of the logo), i also have the same connection with the same corresponding connector pad on the other daughterboard.

If pin 9, most probably the light blue one, isn't supposed to be connected to anything, there should be no trace going to that socket pad... I can't see where and how there could be a short if that socket pad is on its own, disconnected from the whole circuit...

At least, now the effect sounds really good, with more than enough volume on tap. I wonder if it's safe to play with it, despite the pin 9 and pin 4 and 5 situation ?

From a musical point of view, it looks like the jumper alone solved everything...I am really tempted to leave it alone now that everything works.
Could it blow off, electrocute me, or damage the tube because of pins 4, 5 and 9 ?