Hi guys
I could really do with a hint on how to reduce noise in my mu follower design
It sounds really good. But, I have a high level of full spectrum noise. The design is two mu followers sandwiching a passive RIAA, schematic attached. I had previously built a cascode front end common cathode second stage and that was as quiet as a mouse. Removing the first stage tubes shows the noise to be in the second stage and not amplified from the first though that probably has the same issues. There is a slight roll off at high frequency too about a dB. It will be a shame not to get this design to work as it sounds so good.
I have read about loading mu followers can be problematic but not that the effect is noise. I would prefer to keep to all tube and not go hybrid but maybe I have come to the limit of the topology?
I could really do with a hint on how to reduce noise in my mu follower design
It sounds really good. But, I have a high level of full spectrum noise. The design is two mu followers sandwiching a passive RIAA, schematic attached. I had previously built a cascode front end common cathode second stage and that was as quiet as a mouse. Removing the first stage tubes shows the noise to be in the second stage and not amplified from the first though that probably has the same issues. There is a slight roll off at high frequency too about a dB. It will be a shame not to get this design to work as it sounds so good.
I have read about loading mu followers can be problematic but not that the effect is noise. I would prefer to keep to all tube and not go hybrid but maybe I have come to the limit of the topology?
There is no controlled negative feedback so both 12AY7s are running flat out possibly causing pink noise as you describe.
Ensure the second part of each valve is the pre amp side as they have slightly more gain.
The input grid stoppers value is of no use as it must be at least 10k to do anything.
Ensure the second part of each valve is the pre amp side as they have slightly more gain.
The input grid stoppers value is of no use as it must be at least 10k to do anything.
Thanks Jon, I am slightly confused about the gain, I thought it was fixed my the mu of the tubes, in this case the 12ay7’s give a gain of about 44. This is less than the gain I got from the cascode arrangement. Also not sure what you mean about the second part, are you saying take the output from the bottom tube anode and turn it into a SRPP?. I have not seen any design that used feedback unless it was in an active RIAA
What is the rated output of your cartridge?
Moving Magnet and 3.5mV?
Moving Coil and 0.3mV?
Without enough cartridge output, the total gain from cartridge to loudspeaker
is so large that you will hear noise.
Other?
Moving Coils generally require a MC transformer to step up the output.
And those MC transformers can pick up magnetic spray from transformers, switchers, motors, lights, etc.
Using the first stages high impedance plate output to drive the passive RIAA network will require a completely different RIAA network.
A 1 Meg Ohm grid leak resistor on the 2nd stage's grid may be accentuating the 12AY7's noise. But using a lower grid resistor there will also require a re-design of the RIAA network. A high resistance will allow any noisy grid current to become a noisy grid voltage.
Try pulling the first tube, and then putting 100k across the 1 Meg, and see if that reduces the noise.
Moving Magnet and 3.5mV?
Moving Coil and 0.3mV?
Without enough cartridge output, the total gain from cartridge to loudspeaker
is so large that you will hear noise.
Other?
Moving Coils generally require a MC transformer to step up the output.
And those MC transformers can pick up magnetic spray from transformers, switchers, motors, lights, etc.
Using the first stages high impedance plate output to drive the passive RIAA network will require a completely different RIAA network.
A 1 Meg Ohm grid leak resistor on the 2nd stage's grid may be accentuating the 12AY7's noise. But using a lower grid resistor there will also require a re-design of the RIAA network. A high resistance will allow any noisy grid current to become a noisy grid voltage.
Try pulling the first tube, and then putting 100k across the 1 Meg, and see if that reduces the noise.
Last edited:
The 12AY7 low level amplifier service recommends a 100k grid resistor.
Link: https://frank.pocnet.net/sheets/127/1/12AY7.pdf
(Scroll down)
Link: https://frank.pocnet.net/sheets/127/1/12AY7.pdf
(Scroll down)
Try the 12AU7 in the second stage for lower noise. The 33dB should
be ok if you have gain in the line stage, along with the step up.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.