Hi all,
first let me introduce myself, as it's my first post on this forum. I'm a 32 years old electronic engineer from switzerland. I read this forum for a long time, and always found the answers that I was searching, except this time!
I build an active preamp based on a schematic of "Electronique Pratique" magazine. It has a relay-based input selector, preamp stage based on a bipolar transistor, tone control, etc... Nothing really special.
The problem is when the input selector is on a unconnected input I have a big audible noise... When I connect a source (for my tests it was a mp3 player) set on Pause, I have no noise at all. So it means the preamp is quite silent, but the input cables are antennas. Of course I shielded every signal cables, and I have no ground loops.
I made some experiment and saw that a 4k pull-down resistor between input and ground reduce the noise at an acceptable level and does not affect the sound quality or volume (for my ears). But it's quite a small input resistor.
My question is: can a common source handle this input res? My source is not defined, so I don't know the output impedance. What is the minimum recommended value for this input resistor?
Thank you for your help!
Gabriel
first let me introduce myself, as it's my first post on this forum. I'm a 32 years old electronic engineer from switzerland. I read this forum for a long time, and always found the answers that I was searching, except this time!
I build an active preamp based on a schematic of "Electronique Pratique" magazine. It has a relay-based input selector, preamp stage based on a bipolar transistor, tone control, etc... Nothing really special.
The problem is when the input selector is on a unconnected input I have a big audible noise... When I connect a source (for my tests it was a mp3 player) set on Pause, I have no noise at all. So it means the preamp is quite silent, but the input cables are antennas. Of course I shielded every signal cables, and I have no ground loops.
I made some experiment and saw that a 4k pull-down resistor between input and ground reduce the noise at an acceptable level and does not affect the sound quality or volume (for my ears). But it's quite a small input resistor.
My question is: can a common source handle this input res? My source is not defined, so I don't know the output impedance. What is the minimum recommended value for this input resistor?
Thank you for your help!
Gabriel
It would help to see a circuit and some pictures (not sure if you can do that under moderation)
If the input is noisy with no leads connected ? then there has to be a problem with the design/construction etc. 4K is way to low to use as a termination for an input.
What sort of noise is it ? Hum, white noise, or could it be some form of instability rather than "noise" as such. Have you 'scoped the preamp to see if it's going unstable ? Have you tried terminating the input with a small capacitor (say 470pf) to see if that stops it ?
If the input is noisy with no leads connected ? then there has to be a problem with the design/construction etc. 4K is way to low to use as a termination for an input.
What sort of noise is it ? Hum, white noise, or could it be some form of instability rather than "noise" as such. Have you 'scoped the preamp to see if it's going unstable ? Have you tried terminating the input with a small capacitor (say 470pf) to see if that stops it ?
Hi,
Thank you for your reply! It looks like a white noise, it's not a 50Hz ground noise or hum, but I have no way to scope it for the moment (maybe next week at work).
I have noise with no leads connected, so it's not antennas, you're right. I have no 470pF in stock, but I saw that 200pF between input and ground makes not a big difference, but 1nF really reduce this noise. Can I try to increase the value again or will I loose bandwidth?
I also noticed that the noise is acceptable for 3/4 of the volume pot, and exponentially increase in the last 1/4 of the volume control. It is quite "normal" to have noise in this zone, but maybe not at this level... I think I have to rethink the grounding of my circuit, it is build in a wood encolsure, so the box has no faraday effect. The grounding must be perfect...
Cheers
Gabriel
Thank you for your reply! It looks like a white noise, it's not a 50Hz ground noise or hum, but I have no way to scope it for the moment (maybe next week at work).
I have noise with no leads connected, so it's not antennas, you're right. I have no 470pF in stock, but I saw that 200pF between input and ground makes not a big difference, but 1nF really reduce this noise. Can I try to increase the value again or will I loose bandwidth?
I also noticed that the noise is acceptable for 3/4 of the volume pot, and exponentially increase in the last 1/4 of the volume control. It is quite "normal" to have noise in this zone, but maybe not at this level... I think I have to rethink the grounding of my circuit, it is build in a wood encolsure, so the box has no faraday effect. The grounding must be perfect...
Cheers
Gabriel
Grounding will affect hum, but it probably won't affect white noise. That just sounds like normal circuit noise, affected by the design of the front end. Can you post a schematic? It's normal for open inputs to show more noise than terminated ones, and that's one reason people sometimes use shorting plugs on unused inputs.
CH
CH
Are your input grounds tied together and not switched by the relays? Is the pot body grounded as I notice you have a timber enclosure?
If it's white noise it seems to be down to the circuit design (unless it is unstable as mentioned) or some construction error.
We really need to see a circuit.
1nf will probably reduce the bandwidth (how much depends on the source resistance feeding it).
Preamps should be silent, whether discrete of IC.
We really need to see a circuit.
1nf will probably reduce the bandwidth (how much depends on the source resistance feeding it).
Preamps should be silent, whether discrete of IC.
Does it have a 47k input resistor to ground on the input of first first input transistor? If not that's your problem you have an open base.
Cheers George
Last edited:
open input noise attenuation
The famous Dynakit PAS2 preamp shorted the unused inputs to each center to ring with the selector switch. This is fine for tube radios like the Dynakit 1, but may blow the output or run down the batteries on a modern CD player or transistor radio. I found out by trying to balance my phono inputs on my PAS2 with a phase shift oscillator I built out of the GE transistor manual 7th edition, and found the oscillator output potentiometer putting out no voltage when the selector switch was on the wrong thing. I don't imagine 4k will damage a random CD player or pocket radio, but my experience is limited to the 3 or 4 I have around here.
The famous Dynakit PAS2 preamp shorted the unused inputs to each center to ring with the selector switch. This is fine for tube radios like the Dynakit 1, but may blow the output or run down the batteries on a modern CD player or transistor radio. I found out by trying to balance my phono inputs on my PAS2 with a phase shift oscillator I built out of the GE transistor manual 7th edition, and found the oscillator output potentiometer putting out no voltage when the selector switch was on the wrong thing. I don't imagine 4k will damage a random CD player or pocket radio, but my experience is limited to the 3 or 4 I have around here.
Thank you all for your ideas and comments.
First a few answers: yes the input ground are tied together with 1.5mm2 hard copper wire and not switched, only signals are. I must admit that the pot bodies are not grounded, you're right, I'll have to do that. I have no resistor, and for the moment no capacitors, between input and ground...
I attach a schematic of the preamp. As I told it is 100% based on an "Electronic Pratique" schematic that I redraw in EAGLE to order the PCB. I have not enough skills in analogue electronic to modify it, and I only trust the author...
Note also that on of the 4 input has a Phono Preamp and this input is quite silent. So I think the problem should be on the front end of the Preamp. A major Preamp error would cause noise in every case, no?
Cheers
Gabriel
First a few answers: yes the input ground are tied together with 1.5mm2 hard copper wire and not switched, only signals are. I must admit that the pot bodies are not grounded, you're right, I'll have to do that. I have no resistor, and for the moment no capacitors, between input and ground...
I attach a schematic of the preamp. As I told it is 100% based on an "Electronic Pratique" schematic that I redraw in EAGLE to order the PCB. I have not enough skills in analogue electronic to modify it, and I only trust the author...
Note also that on of the 4 input has a Phono Preamp and this input is quite silent. So I think the problem should be on the front end of the Preamp. A major Preamp error would cause noise in every case, no?
Cheers
Gabriel
Attachments
Like I said in my previous post you need a 47k to ground on the input of the preamp, the fact that when you switch to phono verifies this, as the phono output then loads the pre input to ground, as soon as you switch out the phono you have noise again, because the input to the pre has an open base on the first transistor it is not ref to ground via a 47k. Put in a 47k and all your problems will be fixed.
Cheers George
Last edited:
The circuit doesn't look quite right to me around the buffer stage for the tone network.
Should not C4 be driven from the output of the buffer T2 (Collector) ? and the arrangement around TR1 doesn't seem good. Does C5 really feed into the base of T1 ?
Should not C4 be driven from the output of the buffer T2 (Collector) ? and the arrangement around TR1 doesn't seem good. Does C5 really feed into the base of T1 ?
If C5 does feed into the base then the connection to C4 would be on the emitter of T2, not the collector as I stated above 🙂
Oh Dear!!!
http://www.diyaudio.com/forums/anal...ise-unconnected-preamp-input.html#post2137319
Cheers George
http://www.diyaudio.com/forums/anal...ise-unconnected-preamp-input.html#post2137319
Cheers George
- Status
- Not open for further replies.