GM70 SE amp project - Page 3

Rod Coleman · 17th November 2020, 10:38 AM

If you mean filters made with common-mode chokes and "Y" Capacitors (line→safety earth), I think they are not necessary, and are liable to degrade the noise performance.

This kind of filter is produced to address some of the problems of mains→DC coverters, in order to meet the statutory EMC regulations (CISPR22 - EN55022 for IT gear for example). They are intended to reduce the portion of noise from mains→DC converters from getting back into the mains.

But if you have a transformer, with low PRIMARY→SECONDARY capacitance, the insertion loss (noise blocking) is already excellent, especially in the range 1Hz to 10MHz, where filtering requires very large components.

CM choke filtering often uses mH-region inductance values, and this reacts with capacitors to produces big resonance peaks in the impedance plot, often at quite low frequencies.

Meanwhile, the differential inductance of common-mode chokes is excited by the pulse currents of tube amp DC supplies, and generates noise of its own.

For mains isolation and noise control, a good quality mains trafo is needed. If you have one with an interwinding screen, there is no need for input filtering, other than an "X" cap across the Line-Neutral at the input.

gideon1990 · 17th November 2020, 11:58 AM

Okay nice to see my line of thought about this confirmed

audiowize · 17th November 2020, 06:20 PM

Quote:

Originally Posted by Rod Coleman

The problem arises when there is a continuous ground connexion between the DHT amplifier and a DAC, or if the DAC is close enough to the amplifier to pick up EM interference.

This of course assumes that the DAC doesn't also contain a SMPS.

bobikdartanan · 17th November 2020, 06:51 PM

I'll add a little.
Not audible high-frequency noise and high-frequency harmonics of the amplifier also affects the subconscious, which leads to quick fatigue of the listener.
And the quality of the audio system is determined by the listening time without signs of fatigue.

bobikdartanan · 07:53 PM

Sorry, I forgot the main thing:
SE tube amp without NFB, in comparison with transistors amp, usually, but not always

, has a rapidly decaying spectrum of high-frequency harmonics, as well as THD corresponding THD of the ear:
https://citeseerx.ist.psu.edu/viewdo...=rep1&type=pdf
Switching power supply deprives SE tube amp of the main advantage.
as well as diodes in the rectifier:
Diodes

manojrc · 24th November 2020, 01:20 PM

Can someone please help me to get GM70 tube datasheet?

Thank you..

Stephen R · 24th November 2020, 02:01 PM

I used the Meanwell supplies to heat my PP GM70s. Even after full turndown, there was still too much voltage so I burned off some with series resistance and then fed the fil via a CMC, no cap after. It was OK but later (now) I used a VCCS regulator after (Rod's would be another example) and it improved things quite a bit. Would a linear supply be better? I don't know, maybe, but I haven't the funds to find out and the amp sounds plenty good as is.

I would be concerned about the grid resistance for fixed bias in the GM70. 30K is recommended maximum for fixed bias. With 300k you might risk runaway. A 30k grid resistance would be hard for your driver and makes cap coupling very hard. You might need a rethink in this area.

PRR · 24th November 2020, 09:37 PM

Quote:

Originally Posted by manojrc

Can someone please help me to get GM70 tube datasheet?

A modern 1-page sheet:

https://frank.pocnet.net/sheets/084/g/GM70.pdf

Flikoman · 25th November 2020, 05:30 AM

Quote:

Originally Posted by Stephen R

I used the Meanwell supplies to heat my PP GM70s. Even after full turndown, there was still too much voltage so I burned off some with series resistance and then fed the fil via a CMC, no cap after. It was OK but later (now) I used a VCCS regulator after (Rod's would be another example) and it improved things quite a bit. Would a linear supply be better? I don't know, maybe, but I haven't the funds to find out and the amp sounds plenty good as is.

I would be concerned about the grid resistance for fixed bias in the GM70. 30K is recommended maximum for fixed bias. With 300k you might risk runaway. A 30k grid resistance would be hard for your driver and makes cap coupling very hard. You might need a rethink in this area.

Fixed bias via interstage transformer

I use SMPS (24V 4.5A) slightly modified, recapped and RCRC to get 20V. I am very happy with it and it works without issues 5 years now.

deafbykhorns · 25th November 2020, 08:09 PM

Quote:

Originally Posted by manojrc

Can someone please help me to get GM70 tube datasheet?

Thank you..

I have one I can take a picture of if you can read Russian

17th November 2020, 07:50 PM	#25
bobikdartanan diyAudio Member Join Date: May 2016 Location: Israel	Sorry, I forgot the main thing: SE tube amp without NFB, in comparison with transistors amp, usually, but not always, has a rapidly decaying spectrum of high-frequency harmonics, as well as THD corresponding THD of the ear: https://citeseerx.ist.psu.edu/viewdo...=rep1&type=pdf Switching power supply deprives SE tube amp of the main advantage. as well as diodes in the rectifier: Diodes Last edited by bobikdartanan; 17th November 2020 at 07:53 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
GM70 - Mono block project	Nsaran	Tubes / Valves	8	1st October 2019 03:33 AM
GM70 IT PP	claudiomas	Tubes / Valves	58	14th October 2017 11:18 AM
Another GM70	deafbykhorns	Tubes / Valves	2	3rd June 2015 12:36 PM
C3g:GM70:GM70 schematic advice	louischia	Tubes / Valves	24	27th August 2010 05:01 PM
GM70/GM70 parafeed SET?	louischia	Tubes / Valves	10	4th December 2008 04:12 PM

New To Site?	Need Help?
Register to Participate Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

17th November 2020, 10:38 AM	#21
Rod Coleman садовник diyAudio Member Join Date: Jul 2004 Location: Warwickshire UK	If you mean filters made with common-mode chokes and "Y" Capacitors (line→safety earth), I think they are not necessary, and are liable to degrade the noise performance. This kind of filter is produced to address some of the problems of mains→DC coverters, in order to meet the statutory EMC regulations (CISPR22 - EN55022 for IT gear for example). They are intended to reduce the portion of noise from mains→DC converters from getting back into the mains. But if you have a transformer, with low PRIMARY→SECONDARY capacitance, the insertion loss (noise blocking) is already excellent, especially in the range 1Hz to 10MHz, where filtering requires very large components. CM choke filtering often uses mH-region inductance values, and this reacts with capacitors to produces big resonance peaks in the impedance plot, often at quite low frequencies. Meanwhile, the differential inductance of common-mode chokes is excited by the pulse currents of tube amp DC supplies, and generates noise of its own. For mains isolation and noise control, a good quality mains trafo is needed. If you have one with an interwinding screen, there is no need for input filtering, other than an "X" cap across the Line-Neutral at the input.

17th November 2020, 11:58 AM	#22
gideon1990 diyAudio Member Join Date: Sep 2009	Okay nice to see my line of thought about this confirmed

17th November 2020, 06:51 PM	#24
bobikdartanan diyAudio Member Join Date: May 2016 Location: Israel	I'll add a little. Not audible high-frequency noise and high-frequency harmonics of the amplifier also affects the subconscious, which leads to quick fatigue of the listener. And the quality of the audio system is determined by the listening time without signs of fatigue.

24th November 2020, 01:20 PM	#26
manojrc diyAudio Member Join Date: Dec 2009	Can someone please help me to get GM70 tube datasheet? Thank you..

24th November 2020, 02:01 PM	#27
Stephen R diyAudio Member Join Date: Oct 2002 Location: Folkestone	I used the Meanwell supplies to heat my PP GM70s. Even after full turndown, there was still too much voltage so I burned off some with series resistance and then fed the fil via a CMC, no cap after. It was OK but later (now) I used a VCCS regulator after (Rod's would be another example) and it improved things quite a bit. Would a linear supply be better? I don't know, maybe, but I haven't the funds to find out and the amp sounds plenty good as is. I would be concerned about the grid resistance for fixed bias in the GM70. 30K is recommended maximum for fixed bias. With 300k you might risk runaway. A 30k grid resistance would be hard for your driver and makes cap coupling very hard. You might need a rethink in this area.