Hi,
I have put together an attempt at a passive pre-amp using a stepped attenuator and a 3 way input selector.
On the work bench using a voltmeter to measure resistance the signal path seems to be correct with no shorting I can detect. The attenuator is in series with the positive output from the input selector and the positive. Turning it I can see the resistance going from very low to 20k between the positive input and the positive output.
I have used shielded co-ax for the connections, and the stepped attenuator is 20k.
When I connect it to the amp with no source input, there is a buzzing. When the source is connected, the buzzing goes away at full volume. Turning down introduces a milder buzzing, but doesn't attenuate the sound at all after the first notch.
I have tried separate sources cables and power amplifiers, all with the same result.
Do I need a common earth?
Thanks for reading.
I have put together an attempt at a passive pre-amp using a stepped attenuator and a 3 way input selector.
On the work bench using a voltmeter to measure resistance the signal path seems to be correct with no shorting I can detect. The attenuator is in series with the positive output from the input selector and the positive. Turning it I can see the resistance going from very low to 20k between the positive input and the positive output.
I have used shielded co-ax for the connections, and the stepped attenuator is 20k.
When I connect it to the amp with no source input, there is a buzzing. When the source is connected, the buzzing goes away at full volume. Turning down introduces a milder buzzing, but doesn't attenuate the sound at all after the first notch.
I have tried separate sources cables and power amplifiers, all with the same result.
Do I need a common earth?
Thanks for reading.
It seems like capacitive crosstalk into the output wires. Is there a mains wire very close to the attenuator or to the output leads? Is there a shielded enclosure around the whole thing? (Somewhat more exotic possibility: if you have some subcircuit that oscillates parasitically running on a supply with large ripple, you can get the weirdest hum issues, although that sort of thing is more likely in a real amplifier than in a passive attenuator.)
On top of that, there must be something wrong with either the design or the wiring or both, because the volume doesn't go down after the first notch. Is your attenuator a switchable voltage divider (as it should be) or just a switchable series resistor (which would give you very little attenuation)?
On top of that, there must be something wrong with either the design or the wiring or both, because the volume doesn't go down after the first notch. Is your attenuator a switchable voltage divider (as it should be) or just a switchable series resistor (which would give you very little attenuation)?
Last edited:
Thanks.
There is no mains crosstalk. There is a design problem I'm sure.
The attenuator is a voltage divider.
I'm going to try to find a reference design, my understanding was wrong about how to wire up the components.
Is there an easy way to identify which terminals to use as in, out and ground on the voltage divider when they are not marked?
On my one there are two sets of 3 pins. In volume down position, the resistance is 20k between 1 and 2, and no resistance between 2 and 3. In volume up, it is vice versa. There is always 20k between 1 and 3 regardless of volume position. I have concluded from this that in is pin 1, out is pin 2 and ground is pin 3. Is that correct?
There is no mains crosstalk. There is a design problem I'm sure.
The attenuator is a voltage divider.
I'm going to try to find a reference design, my understanding was wrong about how to wire up the components.
Is there an easy way to identify which terminals to use as in, out and ground on the voltage divider when they are not marked?
On my one there are two sets of 3 pins. In volume down position, the resistance is 20k between 1 and 2, and no resistance between 2 and 3. In volume up, it is vice versa. There is always 20k between 1 and 3 regardless of volume position. I have concluded from this that in is pin 1, out is pin 2 and ground is pin 3. Is that correct?
Last edited:
I had that thing with hum reducing to zero at max volume in a passive pre i built a while ago, never did work out why that happens. I rewired it using twisted pairs and made a common earth for all RCA sockets. That worked ok. Hope this helps. If not can you post a diagram of your wiring including connections at the pot.
If your pot has a metal can/body it might help to ground it too. I had to connect the common signal grounds to the alloy case in one i built to stop hum.
If your pot has a metal can/body it might help to ground it too. I had to connect the common signal grounds to the alloy case in one i built to stop hum.
On my one there are two sets of 3 pins. In volume down position, the resistance is 20k between 1 and 2, and no resistance between 2 and 3. In volume up, it is vice versa. There is always 20k between 1 and 3 regardless of volume position. I have concluded from this that in is pin 1, out is pin 2 and ground is pin 3. Is that correct?
That sounds logical (after the edit).
That sounds logical (after the edit).
Damn, you caught me having a blonde moment. I was tired
Davym, thank you. I am going to fix a couple of issues with the grounding, then if it still doesn't work I will upload a circuit diagram
You haven't built a potential divider, you just think you have. Double-check the wiring.jamesd256 said:Turning down introduces a milder buzzing, but doesn't attenuate the sound at all after the first notch.
The input and output signals need to be referenced to the same ground. That is because a point does not have a voltage; only two points can have a voltage between them.Do I need a common earth?
The attenuator behaves as both a RECEIVER and a TRANSMITTER of Signals.
As a Receiver it gets a two wire connection from the Source.
You MUST have that two wire signal connection from Source to Receiver.
As a Transmitter it sends a Signal to the next stage in the system
You MUST have a two wire signal connection from Transmitter to Next Input.
Note that nowhere in any of the preceding have I use "earth" nor the word "ground".
You NEED a TWO WIRE connection to convey the signal between modules.
Your attenuator is one of these modules.
Look again very carefully and identify the two 2wire connections into and out of the attenuator. i.e. 8 wires in total for the 2channels.
As a Receiver it gets a two wire connection from the Source.
You MUST have that two wire signal connection from Source to Receiver.
As a Transmitter it sends a Signal to the next stage in the system
You MUST have a two wire signal connection from Transmitter to Next Input.
Note that nowhere in any of the preceding have I use "earth" nor the word "ground".
You NEED a TWO WIRE connection to convey the signal between modules.
Your attenuator is one of these modules.
Look again very carefully and identify the two 2wire connections into and out of the attenuator. i.e. 8 wires in total for the 2channels.
Last edited:
Thank you AndrewT, DF96.
Very grateful for all the replies. I took this project on thinking I understood the basic signal path, but I didn't understand the principles of the potential divider properly, nor the principles of grounding needed in the design.
I don't completely understand the theoretical points offered, but I now understand I need a common earth rail for the RCA connectors earthed to the chassis, and I am pretty sure I can wire up the potential divider properly.
I think I will try to upload a schematic before I try to solder any more.
Very grateful for all the replies. I took this project on thinking I understood the basic signal path, but I didn't understand the principles of the potential divider properly, nor the principles of grounding needed in the design.
I don't completely understand the theoretical points offered, but I now understand I need a common earth rail for the RCA connectors earthed to the chassis, and I am pretty sure I can wire up the potential divider properly.
I think I will try to upload a schematic before I try to solder any more.
There is a lot to be said for starting off with a simple volume pot (plus selector switch) in a box. That gives the opportunity to learn the basics before trying something more expensive and more likely to go wrong.
We haven't offered any "theoretical points", just some very practical points.
We haven't offered any "theoretical points", just some very practical points.
Yes I have found out before it's hard. Even this was harder than I thought it would be.
I re-wired the attenuator as indicated here, and grounded all the RCA connectors to a common rail connected to the chassis. Now things are better, with no buzzing, but the volume control still isn't right.
At 0 volume (step 0) the signal is greatly, but not completely attenuated. Up one notch, and it jumps quite a bit, then each subsequent step up to full volume decreases attenuation slightly.
Simply put, it's something like this:
(Position,Volume)
0: 20%
1: 60%
2: 65%
...etc
19: 95%
20: 100%
Trying out a few tools to help me produce the schematic. Any suggestions?
Thanks.
I re-wired the attenuator as indicated here, and grounded all the RCA connectors to a common rail connected to the chassis. Now things are better, with no buzzing, but the volume control still isn't right.
At 0 volume (step 0) the signal is greatly, but not completely attenuated. Up one notch, and it jumps quite a bit, then each subsequent step up to full volume decreases attenuation slightly.
Simply put, it's something like this:
(Position,Volume)
0: 20%
1: 60%
2: 65%
...etc
19: 95%
20: 100%
Trying out a few tools to help me produce the schematic. Any suggestions?
Thanks.
Turn the vol pot down to minimum output.
Measure the input impedance/resistance at the input cable plug on both channels. (R1R & R1L)
Measure the output impedance/resistance at the output cable plug. (R2R & R2L)
Turn the vol pot up to maximum output.
Measure the input impedance/resistance at the input cable plug. (R3R & R3L)
Measure the output impedance/resistance at the output cable plug.(R4R & R4L)
Post those eight results.
Measure the input impedance/resistance at the input cable plug on both channels. (R1R & R1L)
Measure the output impedance/resistance at the output cable plug. (R2R & R2L)
Turn the vol pot up to maximum output.
Measure the input impedance/resistance at the input cable plug. (R3R & R3L)
Measure the output impedance/resistance at the output cable plug.(R4R & R4L)
Post those eight results.
Draw it on a sheet of paper. Take a photograph (in focus!!). Post the photo.jamesd256 said:Trying out a few tools to help me produce the schematic. Any suggestions?
Then post a photo of what you actually built.
I'd say your connections at the pot must be wrong. You might have connected the wiper to signal return.
Yes that was it. It's working fine now.
Big amp you once again to all. I really appreciate all the unimpeded input. The volume of help has been great. Time to switch back to listening to music.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Source & Line
- Analog Line Level
- Buzzing passive pre-amp