Hello all, newbie here but frequent contributor to Audiokarma and Audio Asylum under the name Maxamillion (someone already had it here).
I have a question regarding the MOSFET outputs in a low-frequency equalizer that is used with Infinity RSIIb speakers (early 1980s vintage). The equalizer was sold with the speakers and is used to boost the very low frequencies and to help the transition from the two 10" poly woofers to the planar EMIM midranges.
I have extensively modded the equalizer, replacing all the resistors with 1/2W PRP and TX2575 resistors, replacing and bypassing all the old electrolytic caps, upgrading the output coupling caps and wiring, and installing Cree SiC rectifier diodes. The equalizer now sounds MUCH more transparent than previously, but I can still tell when it is in line between my Supratek tube preamp and Gilmore Raptor class D monoblocks. I'm wondering if the original MOSFETS, VN40AD designation, have a more modern equivalent that might sound better. The VN40AD is no longer made, so I'll eventually need to find a replacement if they ever die. I have a few VN40AF which are also obsolete but I suspect they will sound pretty close to the ADs.
Any help from the MOSFET experts is greatly appreciated!
I have a question regarding the MOSFET outputs in a low-frequency equalizer that is used with Infinity RSIIb speakers (early 1980s vintage). The equalizer was sold with the speakers and is used to boost the very low frequencies and to help the transition from the two 10" poly woofers to the planar EMIM midranges.
I have extensively modded the equalizer, replacing all the resistors with 1/2W PRP and TX2575 resistors, replacing and bypassing all the old electrolytic caps, upgrading the output coupling caps and wiring, and installing Cree SiC rectifier diodes. The equalizer now sounds MUCH more transparent than previously, but I can still tell when it is in line between my Supratek tube preamp and Gilmore Raptor class D monoblocks. I'm wondering if the original MOSFETS, VN40AD designation, have a more modern equivalent that might sound better. The VN40AD is no longer made, so I'll eventually need to find a replacement if they ever die. I have a few VN40AF which are also obsolete but I suspect they will sound pretty close to the ADs.
Any help from the MOSFET experts is greatly appreciated!
Hmm, must have been a tough question. Let me re-phrase:
Is there a possibility of improving the sound of the equalizer by replacing the old vertical N-channel enhancement MOSFETs with newer technology? Has there been a leap forward in this technology over the past 25 years, or are the old MOSFETs just as good as the newest ones?
Is there a possibility of improving the sound of the equalizer by replacing the old vertical N-channel enhancement MOSFETs with newer technology? Has there been a leap forward in this technology over the past 25 years, or are the old MOSFETs just as good as the newest ones?
I figured that the difference was only the package from looking at the datasheets. They don't run anywhere near their limits in this application, in fact they don't even have heatsinks on them, so the AF should work fine as a replacement if I need one.
What I'm wondering is whether there are newer n-channel enhancement MOSFETs available which will have improved audio-relevant properties in this application. The entire signal passes through them, so any improvements I can make should be beneficial.
I already replaced the opamps with OPA2107s, and noticed a bit fuller and tighter bass.
What I'm wondering is whether there are newer n-channel enhancement MOSFETs available which will have improved audio-relevant properties in this application. The entire signal passes through them, so any improvements I can make should be beneficial.
I already replaced the opamps with OPA2107s, and noticed a bit fuller and tighter bass.
I asked International Rectifier this question, and their tech guy replied that the IRF4104PBF HexFET would be a good modern replacement for the VN40AD. Looking at the specs it has the same breakdown voltage but the Rds is 1000X lower at the same Vgs.
I'm thinking that lower Rds will be a good thing for performance, correct?
The capacitances are quite a bit higher for the HexFET, however. The input cpacitance is 3000pF for the HexFET vs. 50pF for the VN40AD, the output capacitance is 660pF for the HexFET vs. 50pF, and the reverse transfer capacitance is 380pF for the HexFET vs 10pF, all measured at Vds = 25V, Vgs = 0 and f = 1MHz.
What does the higher capacitance mean in terms of performance as an audio amplifier?
I'm thinking that lower Rds will be a good thing for performance, correct?
The capacitances are quite a bit higher for the HexFET, however. The input cpacitance is 3000pF for the HexFET vs. 50pF for the VN40AD, the output capacitance is 660pF for the HexFET vs. 50pF, and the reverse transfer capacitance is 380pF for the HexFET vs 10pF, all measured at Vds = 25V, Vgs = 0 and f = 1MHz.
What does the higher capacitance mean in terms of performance as an audio amplifier?
Rds(on) is the resistance between drain and source when the device is fully opened, important for switching purposes, Rds(On) pretty much determines how many amps the drain of the device can handle.
Rds(On) is not relevant for audio purposes, as in linear operation.
An IRF610 has stats which are more comparable with your VN40 parts.
An option would be to raise the bias level to 50mA and higher, hence the heatsink comment
Other alternative would be the use of Lateral MOSFETs, Hitachi 2SK213 or higher.
Datasheet of the VN40AD=> http://www.datasheet.org.uk/pdf-datasheets/Databooks-1/Book203-1791.pdf
Sorry I don't get that?
[QUOTEis used to boost the very low frequencies and to help the transition from the two 10" poly woofers to the planar EMIM midranges.][/QUOTE]
Certainly you should hear the presence of the equalizer? That's its purpose, no? I mean that gear actively manipulates the frequency response of the system, so how could it be otherwise?
Somehow I miss what all the others seem to understand.
Have fun, Hannes
[QUOTEis used to boost the very low frequencies and to help the transition from the two 10" poly woofers to the planar EMIM midranges.][/QUOTE]
Certainly you should hear the presence of the equalizer? That's its purpose, no? I mean that gear actively manipulates the frequency response of the system, so how could it be otherwise?
Somehow I miss what all the others seem to understand.
Have fun, Hannes
Certainly you should hear the presence of the equalizer? That's its purpose, no? I mean that gear actively manipulates the frequency response of the system, so how could it be otherwise?
Somehow I miss what all the others seem to understand.
Have fun, Hannes[/QUOTE]
Yes, perhaps I worded that unclearly. The low frequency equalizer should only affect frequencies below 1KHz, and ideally be transparent to any frequencies above that, which should pass through it unchanged. I notice a difference in the high frequencies, a kind of slight hardness or crystallinity, that I would like to eliminate. I have already changed the output capacitors to much better quality, which has helped, but I'm wondering if using a better MOSFET might get me all the way there.
no.
It stays the same size 115k.
crop the schematic to fit within the the Forum's limits and post that as an attachment.
Those two MOSFETs form a Single Ended Class A buffer (upper is a Source Follower, the lower one is CCS running on about 25mA of Id).
If I were you I'd not lose my sleep over VN40 (nothing really special about that part in that circuit). I'd rather make a nice buffer with better parts (2SK2013 comes to mind) which would surely be an improvement to your EQ.
A Zero Tempco of 300mA at a Vgs of 4V isn't that bad, sort of reminds me of the Superdupertex VN13**/VP13** series.
(even a used-to-be favored by some of the big boyz like CH)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.