ODS #183 PCB build.

[jazzbass]

Hi Jan,

thanks to you I have also discovered that layout and wiring diagram are in disagreement, the layout shows that the NFB is connected to the 4 Ohm transformer output while the diagram shows the same connection to the 8 Ohm output. I swear I hadn't noticed it yet.

I will certainly leave the NFB connected to the 4 Ohm output.

I confirm that connecting the motherboard with everything attached to the chassis is a nightmare. I started by first soldering the wires to the external components of the board but I had to give up because many connection points are not at the edges of the board but very far from the edges.
After a moment of confusion I decided to stop and dedicate myself to the wooden part that will contain the chassis .... much more relaxing

Franco

[digi2t]

That's interesting. I saw the connection to the 4 ohm tap as well, but just connected to the 8 ohm tap out of habit of seeing it done this way on the Traynors. From what I've read, it doesn't really make much difference which tap the feedback is connected to. Mine seems to be working quite well connected to the 8 ohm tap. As for the schematic that I attached in my last post, the connection to the 8 ohm lug is a reflection of what I did on mine, and NOT what was indicated originally. I apologize if that may have created any confusion.

Franco, that is a sweet looking build. I thought about doing a natural wood cab myself, but being winter, I don't feel like working outside on it, so I commissioned a cab builder to make my cabs for both units. The second one is for my buddy in the UK, so I need to get it cabbed sooner than later.

I agree on the ground points to some degree, but that's probably a byproduct of making the board multi-purpose. In the end, it wasn't really a huge issue. For the 183 build, I used a jumper to link the two grounds in the 12v section together, and then one wire from there to the ground point. I used the 124 layout to transpose ground locations from the PCB around the chassis. For places where multiple ground wires converge, I simply used multiple ground lugs to allow for adequate solder connections to that point. These are the lugs I used;



As for the bias cap, again, a cue I took from Traynor in that respect. Besides... I didn't want to see that cap hanging off the board like that.

Speaking of caps, pay close attention to the correction I made on C23 in the 12v section. In the AA layout, the + of the cap is connected to the wrong side of the diode.

As for wiring, I did whatever wiring that is within the chassis first. Then I wired leads to all the points of the PCB, just eyeball cutting them a bit longer than what is needed. I then dropped the "Medusa" into the chassis, and went around connecting each lead to it's respective point. I finished with the flying heater wires.

All in all, pretty adventurous for a first (and second!) build, especially back to back. Kept me busy for sure, but they turned out great, so I'm happy. They sound awesome to boot. I'm already planning a 6L6 high plate/skyline HRM build, but on a Luigi board (HRD2ODS). I have the board, a chassis and tubes are on their way, and the transfos from Edcor arrived the other day. Just need to order all the internals now. As for the tubes, as per my affinity to the 6CA7 over the EL34, I also have a preference for the 7581A over the 6L6, so that's what I will be using in it.

Funny story time; I'm getting ready to fire up the second build. Tubes are out, bias pot at max negative voltage, current limiter is plugged in, all set. Plug into power, throw the switch... nothing from the limiter. Hook up the DMM, start probing, zero voltage. Disconnect, start looking over all my connections. I'm a bit paranoid now because the first unit I tested and ran on 120v, but then had to reset the wiring to 240v, since it's going to the UK. Anyway, I double check all my connections. Nothing shady there. I power up again, and... nothing.

OK....

I sit there for what seems like a solid 5 minutes just staring at it, when suddenly it dawns on me... THE FUSE ISN'T IN! Throw a fuse in, power up again. Looking at the limit bulb, I don't see that telltale initial brightness and then dimming of the caps charging. Scratching my head again. Power down, check the bulb, nope, bulb is good. Try it again. This time, I'm sitting in a different position, casting a shadow on the bulb. I see a VERY dim glow. I mean, really dim. The lighting in my shop is so bright, It's cancelling out the bulb. Double check with the DMM, yup, voltages are there.

At that point, I decided it was a good time to take a break.

[jazzbass]

Good morning digi2t and ijedouglas, (at least in Italy it's morning)

not without difficulty I finished the assembly of the components:

I have at home some old stand-up with English thread, I had to redo the thread with the result of having to use bigger screws that did not go through the holes of the motherboard that I had to widen.

I turned on the amplifier with the 100 watt lamp in series to check if there were shorts: everything ok <I remembered that someone had not put the fuses after the first failed ignition attempt: .

Switched on with 220 volts: loud noise on the speakers. Evidently the connection to the EL34 valves of the output transformer was reversed.

When the wires of the output transformer are reversed, the amplifier comes to life and has a clean very "important". Great

Quite noticeable mass noise: I have to check the mass points and rely on the original scheme.

The final tubes become very bright red: Bias circuit to be reviewed because the minimum cathode voltage is 46 mv.

I ordered the power transformer with the relays output 6 volt without thinking about the differences between the Ceriatone scheme and the reference scheme that Andrea Antonello had sent. In the diagram the output is 12 volts while in the layout it is 6 volts. Ceriatone also reports 6 volts .... I ordered Omron 12 volt relays . I think I will have to redo the power supply circuit using a voltage doubling scheme .... or replace the relays but it is an almost impossible undertaking since they are soldered to the motherboard directly.

Obviously neither the OD circuit nor the FET circuit work, about which I wanted to complicate my life further. I didn't use the classic boost that takes the mids and lows out of the circuit but use the FET (which is hardly used) as a booster. I still don't know if it works due to the wrong voltage on the relays but I have faith that it is a good choice. It is clear that I don't like simple things and that I love to complicate my life but ....

If there are any recommendations on the circuit for the relays they are welcome

Franco

[jazzbass]

Hello everybody,

I suspect that, in addition to the error reported by digi2t -"Speaking of caps, pay close attention to the correction I made on C23 in the 12v section. In the AA layout, the + of the cap is connected to the wrong side of the diode."- there is another problem with the diodes. If we number the diodes from left to right 1,2,3 and with 4 the diode on the right side of the capacitor C23 it can be seen that the diodes 1 and 3 are connected in parallel, furthermore the diodes 1,2,3 they are connected to the same wire of the voltage 0-50 volts. Now, calmly, I will squeeze my brain to make sense of my observations

I tried to bring the bias values that were too high to normal by increasing the resistance from the bias pot to mass but I think I will have to replace the 10K pot with a 50K one.

Franco

[ijedouglas]

Hi Franco,

No need to change the bias pot to 50K. You can just change the 3K9 resistor in the bias circuit. Depending on your transformer, you may need to tweak that resistor to get a useable range on the bias pot. I think I used a 3K3 in my last 183 build

Ian

[digi2t]

Hello everybody,

I suspect that, in addition to the error reported by digi2t -"Speaking of caps, pay close attention to the correction I made on C23 in the 12v section. In the AA layout, the + of the cap is connected to the wrong side of the diode."- there is another problem with the diodes. If we number the diodes from left to right 1,2,3 and with 4 the diode on the right side of the capacitor C23 it can be seen that the diodes 1 and 3 are connected in parallel, furthermore the diodes 1,2,3 they are connected to the same wire of the voltage 0-50 volts. Now, calmly, I will squeeze my brain to make sense of my observations

I tried to bring the bias values that were too high to normal by increasing the resistance from the bias pot to mass but I think I will have to replace the 10K pot with a 50K one.

Franco

I don't see an issue with the diodes. They make a proper bridge rectifier. The image is deceiving by the way the traces intersect with the diode connections, particularly for the upper right most diode. Otherwise, in reality, they are oriented properly. I've highlighted the paths in this image for clarity. Yellow/orange for AC, red for positive, and green for ground. The aforementioned diode looks to connect with the yellow line, but it does not. It connects to the orange pad;

12v power.jpg

This section is not connected to the bias supply at all.

As for the bias being too high, I encountered the same problem with the second build, not being able to get my bias below 41mA. I simply increased the 22K resistor to ground to 24K and it was enough to give me adequate adjustment. 27K would have been even better, but a 24K 1/2w was all I had on hand at the time. It was much easier to replace being directly between the pot and ground tab, rather than pulling the 3K9 off the board.

I used a separate Hammond 266GD12 series xfo for the relays (12/6v - 0.7/1.4A), using the Omron G5V-2 series relays. I opted for the H1 sub-model, which is the high sensitivity. While the contact rating is a bit less than the standard unit, it's still plenty beefy for this application. The H1 also offers a higher mechanical durability, as well as a 180% coil voltage tolerance over the 120% of the standard unit. This means that even if your 7812 decides to take a powder, the relay will still operate on up to 21vdc. I tried them directly after the bridge, bypassing the regulator, and they worked fine. Also a higher temperature tolerance as well. Pretty well bulletproof.

[jazzbass]

Hello everybody,

I suspect that, in addition to the error reported by digi2t -"Speaking of caps, pay close attention to the correction I made on C23 in the 12v section. In the AA layout, the + of the cap is connected to the wrong side of the diode."- there is another problem with the diodes. If we number the diodes from left to right 1,2,3 and with 4 the diode on the right side of the capacitor C23 it can be seen that the diodes 1 and 3 are connected in parallel, furthermore the diodes 1,2,3 they are connected to the same wire of the voltage 0-50 volts. Now, calmly, I will squeeze my brain to make sense of my observations

I tried to bring the bias values that were too high to normal by increasing the resistance from the bias pot to mass but I think I will have to replace the 10K pot with a 50K one.

Franco

I don't see an issue with the diodes. They make a proper bridge rectifier. The image is deceiving by the way the traces intersect with the diode connections, particularly for the upper right most diode. Otherwise, in reality, they are oriented properly. I've highlighted the paths in this image for clarity. Yellow/orange for AC, red for positive, and green for ground. The aforementioned diode looks to connect with the yellow line, but it does not. It connects to the orange pad;

12v power.jpg

This section is not connected to the bias supply at all.

As for the bias being too high, I encountered the same problem with the second build, not being able to get my bias below 41mA. I simply increased the 22K resistor to ground to 24K and it was enough to give me adequate adjustment. 27K would have been even better, but a 24K 1/2w was all I had on hand at the time. It was much easier to replace being directly between the pot and ground tab, rather than pulling the 3K9 off the board.

I used a separate Hammond 266GD12 series xfo for the relays (12/6v - 0.7/1.4A), using the Omron G5V-2 series relays. I opted for the H1 sub-model, which is the high sensitivity. While the contact rating is a bit less than the standard unit, it's still plenty beefy for this application. The H1 also offers a higher mechanical durability, as well as a 180% coil voltage tolerance over the 120% of the standard unit. This means that even if your 7812 decides to take a powder, the relay will still operate on up to 21vdc. I tried them directly after the bridge, bypassing the regulator, and they worked fine. Also a higher temperature tolerance as well. Pretty well bulletproof.

Hi digi2t,

I followed the tracks underside of the mother board and you are absolutely right. I measured the voltage after the bridge and it is 8 volts, too little for 12 volt powered relays. Since I have chosen a vertical PT I have the necessary space inside the chassis for a 220-12v transformer.

Done and the relays work after I corrected another error but don't tell anyone .... I have four ICs including three voltage regulators and a mos-fet, guess which one I mounted ?.

Replaced the resistor from the grounded bias pot with a 27K one, all right.

Now I have to review the mass points, I have an annoying background noise certainly due to a wrong mass.

Tomorrow is a pre-holiday and I hope to find the time to continue the research.

Thank you

Franco

[jazzbass]

Good morning digi2t,

I solved the problem of relays by feeding them with a small 220/12 Volt transformer, now the OD and Boost work.

To bring the Bias current to the right values I used a 33K resistor in series with the 10K potentiometer.

I have to solve the background hum problem that is almost certainly caused by mass problems. Certainly the arrangement of the components in the motherboard does not help and I think this is the most difficult problem to solve. Have you had any problems with the masses in your two amps ?. Fortunately I can compare different schemes, the Ceriatone layout and the Tony Albany layout.

Thank you

Franco

[jazzbass]

Hi everybody,

done, it was a bad solder done a little too quickly and in a hard-to-access part of the amplifier.

Franco

[ijedouglas]

Hi Franco,

Great news.

For future reference, what are you referring to when you say "mass/masses"?

Regards

Ian

[digi2t]

In Europe "earth" or "mass" means what we call ground.

In building mine, my biggest worry in the regard to ground was the anodized chassis. I don't trust the anodizing to be conductive enough, so I took the time to buff off the coating at all ground points, as well as all points where pot bodies, transfo mounting holes, and jacks make contact with the chassis.

I'm glad to hear that you have resolved your issue.

[jazzbass]

hello digi2t, hello ijedouglas,

yes, "massa" or "terra" or "earth" is the equivalent word for the English term ground, if I make a statistic of the errors to be corrected when I finish the construction of an amplifier a bad or forgotten earth is clearly at the top of the chart.

Now the amp is silent as a dead mouse.

Now it's time to do the tuning ..... # 183 sounds good but I have to fix a few things in order to play at maximum.

First: the bias adjustment has different effects on the two EL34s, there are 5.8 mV of difference. I adjusted the bias for an average value of 32 mV. It is likely that the two tubes (I removed them from a Trainwrech Express clone) have aged differently.

Second: the PT primary has 0-220-240 Volt inputs and I connected the 0-240 Volt terminals to the wall socket (220 volts). The secondary of the power transformer is 345-0-345 but after the rectifier diodes I have a voltage B + of 486 Volt instead of 465 Volt. B + 2 is 486 Volts instead of 462 Volts. B + 3 is 455 Volts instead of 441 Volts. B + 4 is 380 Volts instead of 324 Volts. B + 5 is 374 Volts instead of 316 Volts. I should probably change the PSDropping resistors but with what values?

How should I adjust the OD Trim? for what voltage values?
For what voltage values I have to adjust the Trim between R19 and R20 ?

Thanks for your help.

Franco

[ijedouglas]

Franco,

Thanks for the explanation.

Can you rewire your PT to use the 220 primary? Just out of interest, what are your plate voltages?

Set the OD trimmer for 22K to ground as on the layout and then tweak by ear. As you turn it up, you will notice that the sound gets thicker. While it does increase the amount of overdrive, I find it is more about setting the frequency response of the OD channel.

[jazzbass]

Franco,

Thanks for the explanation.

Can you rewire your PT to use the 220 primary? Just out of interest, what are your plate voltages?

Set the OD trimmer for 22K to ground as on the layout and then tweak by ear. As you turn it up, you will notice that the sound gets thicker. While it does increase the amount of overdrive, I find it is more about setting the frequency response of the OD channel.

Hi ijedouglas,

Yes, I can use the 220 volt input by replacing the 240 Volt one, I thought about it but I couldn't give myself a reason and I preferred to do a search on Amp Garage on the PS Resistors String but I didn't find the articles I had already studied few months ago.

Thanks for the tip on the OD Trimmer, now I'll try and write some progress.

Franco

[ijedouglas]

Franco,

Thanks for the explanation.

Can you rewire your PT to use the 220 primary? Just out of interest, what are your plate voltages?

Set the OD trimmer for 22K to ground as on the layout and then tweak by ear. As you turn it up, you will notice that the sound gets thicker. While it does increase the amount of overdrive, I find it is more about setting the frequency response of the OD channel.

Hi ijedouglas,

Yes, I can use the 220 volt input by replacing the 240 Volt one, I thought about it but I couldn't give myself a reason and I preferred to do a search on Amp Garage on the PS Resistors String but I didn't find the articles I had already studied few months ago.

Thanks for the tip on the OD Trimmer, now I'll try and write some progress.

Franco

Franco,

I would highly recommend changing to the 220 tap instead of the PS dropping string. If you can get there without changing the dropping string, I would do it. Each time I have changed the dropping string, I noticed a larger shift in feel than voltages. On my first 183 build, I tried to match the layout voltages and dropped the screen node down to 1K5. While my voltages matched, the amp felt very stiff and less harmonically rich. I started raising the value of the resistors and finally ended up with 3K resistor Mr Dumble used. I have found that tube rolling got me a bigger change in plate voltages than changing the dropping string.