Help Troubleshooting volume loss PLEASE

[Khron]

the 1/4"-connector could be used as a "line-level" input and the XLR-connector could be used as a "mic-level" input. Although such a use would be possibly considered to be "rare", it is nonetheless.....a possibility to accomplish!!!

Not rare at all - in fact, apart from most MOTU gear (with specifically designated inputs such as the Ultralite the OP has), and certain TC Electronic products, damn near ALL audio interfaces with XLR-combo input connectors have the XLR dedicated to mic-level and the TRS to line-level (going into the mic preamp via a resistive pad). Combo inputs that can also act as instrument inputs as well, have that side of things switchable (either physically or via respective the control software).

That's the case at least with Focusrite, Presonus, M-Audio, Alesis etc.

 

[sonolink]

Not rare at all - in fact, apart from most MOTU gear (with specifically designated inputs such as the Ultralite the OP has), and certain TC Electronic products, damn near ALL audio interfaces with XLR-combo input connectors have the XLR dedicated to mic-level and the TRS to line-level (going into the mic preamp via a resistive pad). Combo inputs that can also act as instrument inputs as well, have that side of things switchable (either physically or via respective the control software).

That's the case at least with Focusrite, Presonus, M-Audio, Alesis etc.

Absolutely, and that's probably why I assumed the MOTU was like that (although it was clearly labelled on the front panel

Khron, do you think it's worth having the balanced output on this pedal, considering that the idea behind it, is to avoid carrying an amp for a recording or a gig, and being able to plug directly into a mixer or soundcard?

Cheers
Sono

 

[Khron]

Khron, do you think it's worth having the balanced output on this pedal, considering that the idea behind it, is to avoid carrying an amp for a recording or a gig, and being able to plug directly into a mixer or soundcard?

Why wouldn't it be? Should be great fun to hear just how much more noise you'd get and how huge a (possible) ground loop could form, with the pedal on stage, and a single-ended cable going at the very least 40-50ft / 10-15m or so, all the way back to the mixer...

Just keep in mind to add a ground-lift switch for the output XLR.

 

[sonolink]

Why wouldn't it be? Should be great fun to hear just how much more noise you'd get and how huge a (possible) ground loop could form, with the pedal on stage, and a single-ended cable going at the very least 40-50ft / 10-15m or so, all the way back to the mixer...

Just keep in mind to add a ground-lift switch for the output XLR.

I'm not sure I explained myself well. What I meant to ask was if you reckon it's a good idea to have an inbuilt DI box output in the pedal or it's not worth it since an external DI box can be used.

This build is quite tight in a Hammond 1590BB but I don't want to use a bigger enclosure and at the same time I'm trying to cut on everything I can to gain space.

For instance I have to check how useful the Cab Volume pot is, considering there's also a Gain and a Master pot. If it's not really useful I'd rather put a fixed resistor or directly eliminate the pot to gain space inside the enclosure.

Initially I would have used a transformer for the DI out, something like a 13k:600 but it's also bulky and more expensive.

The DI out can be done with a TRS connector but ideally you would use an XLR, again meaning a more bulky connector.

So I'm trying to get around all these considerations and was just asking fpr advice and opinions. Sorry if it sounded like a dumb question.

Cheers

Sono

 

[MidnightArrakis]

I don't want to use a bigger enclosure and at the same time I'm trying to cut on everything I can to gain space.

[I don't want to use a bigger enclosure and at the same time I'm trying to cut on everything I can to gain space] -- This might come across as being a -- bummer -- for you to read, but.....your reasoning is one of the reasons why things are being designed with "Surface-Mount" components. Whatever circuitry that you have, minus the "electro-mechanical" components (most switches, transformers, connectors, etc.) can all be designed onto a PCB significantly smaller than anything you could come up with using "Thru-Hole" components. Just sayin'.....

/

 

[sonolink]

Thanks for pointing that out

 

[MidnightArrakis]

At the same time, use crocodile-clips from your multimeter to monitor the HV DC supply.

[use crocodile-clips from your multimeter] -- "Alligator-clips" are friendlier!!! More widely available, too!!!

/

 

[MidnightArrakis]

This build is quite tight in a Hammond 1590BB but I don't want to use a bigger enclosure and at the same time I'm trying to cut on everything I can to gain space.

>> Just for your reference:




And.....you built THIS ↓↓↓↓ schematic into THIS enclosure ↑↑↑↑, right???



Just my personal opinion here, OK???.....But, from what I see here looks like some rather -- iffy -- solder-joints:




And.....I'm also guessing that this is some purchased PCB from either a member of this forum or from some other DIY website. Regardless, in either case, the -- annular ring -- of these pads are what I would consider to be "too small" for being hand-soldered. Wave-soldered??? No problem. But, "hand-soldered"??? The "annular rings" should be a bit larger, while also being soldered with a bit more "finesse"!!! The general appearance of these solder joints looks to be a bit "rough", which could also possibly be an underlying cause of some of your "difficult to locate" noise problems. As an example, the solder joint of the anode on the bottom diode looks a bit "iffy" to me. While it could actually be perfectly fine in reality, I would still go ahead and touch it up some.

The three connections shown above of the pot solder joints also all look rather dubious at best. They all could possibly use some "warming up" just to make sure everything is all connected nice and snug!!! A possible "cold solder joint" on one of those connections could be causing some of your "scratchiness" by an ever-so-minute (pronounced: meye - noot, in this case) rapidly intermittent connection as you rotate the control. I once actually encountered this with a brand-new $2,500 compressor from a well-known brand-name. All it took was for me to first track-down the source of my problem and then all of a few seconds to reheat the solder joint on the pot having an issue. PROBLEM SOLVED!!!

/

 

[sonolink]

And.....you built THIS ↓↓↓↓ schematic into THIS enclosure ↑↑↑↑, right???

View attachment 127715

That's right, yes.

And.....I'm also guessing that this is some purchased PCB from either a member of this forum or from some other DIY website.

Actually no. I did the layout and got the board from JLCPCB

Regardless, in either case, the -- annular ring -- of these pads are what I would consider to be "too small" for being hand-soldered. Wave-soldered??? No problem. But, "hand-soldered"??? The "annular rings" should be a bit larger, while also being soldered with a bit more "finesse"!!! The general appearance of these solder joints looks to be a bit "rough", which could also possibly be an underlying cause of some of your "difficult to locate" noise problems. As an example, the solder joint of the anode on the bottom diode looks a bit "iffy" to me. While it could actually be perfectly fine in reality, I would still go ahead and touch it up some.

The three connections shown above of the pot solder joints also all look rather dubious at best. They all could possibly use some "warming up" just to make sure everything is all connected nice and snug!!! A possible "cold solder joint" on one of those connections could be causing some of your "scratchiness" by an ever-so-minute (pronounced: meye - noot, in this case) rapidly intermittent connection as you rotate the control. I once actually encountered this with a brand-new $2,500 compressor from a well-known brand-name. All it took was for me to first track-down the source of my problem and then all of a few seconds to reheat the solder joint on the pot having an issue. PROBLEM SOLVED!!!

/

You're right. When I did the second board I took care of soldering every spot carefully and the noise was gone. So probably that was the cause.

 

[sonolink]

Hi guys,

If I'd want to use a transformer for the balanced out could you tell me if this one would be suitable please?

https://es.aliexpress.com/item/4000...50859Dlu&utparam-url=scene:search|query_from:


Connected backwards like this:



Would that work and be ok?

Thanks a lot
Cheers
Sono

 

[ccaudle]

If I'd want to use a transformer for the balanced out could you tell me if this one would be suitable please?

Just for guitar? The frequency response listed on the purchase page is already dropping by 100Hz, so not sure I would want to use it for bass.
Also no indication what level the transformer was tested at, distortion at that tested level, etc.
My general feeling is that you should only purchase a product from a vendor which does not seem to have any expertise in the product sold if you are willing to discard the product if it turns out to be junk. It could work OK, but there is no way to be sure from the limited amount of information on the aliexpress page.

Do you want the balanced output to drop the level? Using that transformer will drop the level by 12dB at the balanced output. I'm just making sure that is an explicit choice. Since the balanced output is driven by an op-amp you do not necessarily have to drop the level if dropping the level does not fit some explicit goal. You could either leave the output as directly connected, or use a 1:1 output transformer if you don't need that 12dB level drop.

 

[sonolink]

Just for guitar? The frequency response listed on the purchase page is already dropping by 100Hz, so not sure I would want to use it for bass.
Also no indication what level the transformer was tested at, distortion at that tested level, etc.
My general feeling is that you should only purchase a product from a vendor which does not seem to have any expertise in the product sold if you are willing to discard the product if it turns out to be junk. It could work OK, but there is no way to be sure from the limited amount of information on the aliexpress page.

I agree with you. I just thought that for that price it would be worth a try...

Do you want the balanced output to drop the level? Using that transformer will drop the level by 12dB at the balanced output. I'm just making sure that is an explicit choice. Since the balanced output is driven by an op-amp you do not necessarily have to drop the level if dropping the level does not fit some explicit goal. You could either leave the output as directly connected, or use a 1:1 output transformer if you don't need that 12dB level drop.

My idea was to leave out the balanced output and use a transformer to create one. Something like this:

My question was more about if the connection and the transformer ratio were correct to obtain a DI Out to plug it into a mic input on a desk for instance.

Thanks a lot for your help
Cheers
Sono

 

[ccaudle]

My question was more about if the connection and the transformer ratio were correct to obtain a DI Out to plug it into a mic input on a desk for instance.

Yes to the transformer ratio, it should work OK for that.

Regarding the connection, the 10K resistors to gnd compromise the isolation, so connecting the transformer directly to pins 2 and 3 is typical.
By compromise the isolation I mean that if you decide to lift the shield to chassis connection (please don't actually connect the shield switch directly to circuit ground node as you have drawn) then the 10K resistors to gnd still link the receiving device to the pedal gnd node, which negates the point of having a ground lift switch.

 

[sonolink]

Yes to the transformer ratio, it should work OK for that.

Great

Regarding the connection, the 10K resistors to gnd compromise the isolation, so connecting the transformer directly to pins 2 and 3 is typical.

Of course!! I'm sorry, I must've drawn that before my morning coffee....
I don't need C14, C17, R26, and R27 anymore to protect against incoming phantom power since I'm using a transformer.

BUT, how can I find out if that transformer will be enough protection? Voltage? Power?

By compromise the isolation I mean that if you decide to lift the shield to chassis connection (please don't actually connect the shield switch directly to circuit ground node as you have drawn) then the 10K resistors to gnd still link the receiving device to the pedal gnd node, which negates the point of having a ground lift switch.

You're right....

Would this be the right wiring?




Cheers
Sono

 

[ccaudle]

Would this be the right wiring?

For the signal yes.

But the way that schematic is drawn makes it look like you are making a switch to choose whether you intentionally inject cable shield noise currents into your signal path or not:


Don't take the term "ground lift" too literally, it is still a shield connection. Maybe think of it as "shield lift" and it will help.

 

[sonolink]

For the signal yes.

But the way that schematic is drawn makes it look like you are making a switch to choose whether you intentionally inject cable shield noise currents into your signal path or not:
View attachment 128546

Don't take the term "ground lift" too literally, it is still a shield connection. Maybe think of it as "shield lift" and it will help.

Sorry, I don't completely follow you.....I took the wiring from Elliott's web. Figure 2:

https://sound-au.com/project35.htm

The XLR connector is the OUTPUT...
Where is my mistake?

Cheers
Sono

 

[sonolink]

Bump.....

 

[Graham Nystrom]

Shield of a wire connects to the CHASSIS or CASE. Which is different that GND. The shield of the wire does NOT connect to GND.

CHASSIS/CASE connects to ground at one place only and that is close to the AC mains as possible.

Does that clear it up?

The drawing in your link is a little differnt than yours. Where you connect Chassis and GND, you often run that parallel through a resistor and capacitor as shown in the link drawing. Chassis always eventually meets ground, and thats why its shown in the drawing.

 

[sonolink]

Thanks for your reply but now I'm even more confused .
I'm sure this is just a language barrier.

Shield of a wire connects to the CHASSIS or CASE. Which is different that GND. The shield of the wire does NOT connect to GND.

What? How can the wire's shield work if not connected to GND? Where do you connect it then?

CHASSIS/CASE connects to ground at one place only and that is close to the AC mains as possible.

I agree that Chassis/case connects to GND at one place only, only there's no AC mains here. Since Chassis/Case connects to GND I understand that they are the same only their purpose changes. Isn't that right?

The drawing in your link is a little differnt than yours. Where you connect Chassis and GND, you often run that parallel through a resistor and capacitor as shown in the link drawing. Chassis always eventually meets ground, and thats why its shown in the drawing.

I'm sorry, I didn't redraw the resistor/capacitor. BUT My initial drawing in post #92 shows the resistor/cap net. What I still don't understand is what ccaudle meant when he said:

But the way that schematic is drawn makes it look like you are making a switch to choose whether you intentionally inject cable shield noise currents into your signal path or not:
View attachment 128546

Don't take the term "ground lift" too literally, it is still a shield connection. Maybe think of it as "shield lift" and it will help.

Did he say that because the resistor/cap net was missing? If so, sorry for the confusion

BTW, what's the use of the resistor/cap?

Thanks for your time and help
Cheers
Sono

 

[ccaudle]

Sorry, I don't completely follow you..

Sorry for the delay, I got swamped with something at work for a couple of days.
I just meant that pin 1 problem is so recurring, that schematics should be very explicit where things connect together. The way that ElliotSound circuit is drawn, the naive way to follow the connection diagram would be to connect the circuit reference (aka "ground") node directly to the shield connection, which is going to cause noise problems.

This is a more explicit (I would argue the correct) way to notate the connections:

I'm sure this is just a language barrier.

It is in the sense that many people use terminology very loosely and use the word "ground" to refer to many different things. That is why I like to use explicitly accurate words like "shield connection," "shield conductor," "chassis connection," "circuit reference node" in contexts where it is not overly cumbersome and doesn't just make a muddle of the text.

How can the wire's shield work if not connected to GND?

You have mixed up the concept of circuit reference node and shield in your head. You first have to correct that in order to think correctly about where various conductors connect together.
A shield works perfectly well with no connection to the circuit if the circuit does not have to connect to the outside world. That is a very, very rare condition to be met (the only situation I can think of off the top of my head is battery powered instrumentation trying to measure something like x-ray intensity, or acoustic noise that you want to completely enclose for EMI shielding).
In our world of audio equipment you will always be connecting different equipment together, so you want to minimize the potential difference between the shield and the conductors travelling inside the shield to reduce capacitive coupling of shield noise, and you need a way to keep the common mode voltage between different pieces of equipment within reasonable limits, so you connect the circuit 0V reference node to the shielding enclosure at one point so that their voltage potentials are tied together.
By connecting only at one point you do not create a path for current to flow through the audio circuit due to voltage differences at different points on the shielding enclosure.

only there's no AC mains here

[regarding advice to connect at the AC mains entrance]
So you just pick a reasonable point. Possibly where the external power supply enters is a convenient place.

Since Chassis/Case connects to GND I understand that they are the same

What do you mean by stating "they are the same?" The circuit reference is for audio signal return currents and being the "0" reference for the audio signal, the shield is for forming a conductive surface around the circuit so that external EM fields generate currents in the conductive enclosure and not in the audio circuitry.
Those are different things, so I would say that no, they are not the same. Thinking that they are the same and that they can be connected together at any random point without regard for where currents flow and where those currents generate voltages across the conductor resistance has been a perpetual source of noise problems in otherwise decent gear.

what's the use of the resistor/cap?

The resistor-capacitor combination is to reduce the current at audio frequencies if there is an inadvertent dual connection formed, while attempting to have a somewhat low impedance at radio frequencies.

Note that having a switch in the shield path makes it mostly ineffective at radio frequencies. My preference is to correctly connect the shields with as low inductance a connection as possible, and if you find that you need to break the shield path either use a cable with shield connected at the transmit end only, or use a female-to-male XLR adapter with pin 1 disconnected internally at the receiving equipment end.