Ale,
What do you think of the mu-follower gyrator as the load for a single 6e5p to drive an 833 at 1500V and 250mA, in heavy A2? I'm working up an 833 amp and was instead thinking of using a CCS on top of the 6e5p followed by an AOT1N60 MOSFET with separate +/- supplies in order to ensure enough drive. Do you think the simple depletion mode mosfet gyrator could do the job? I'm a little doubtful...
What do you think of the mu-follower gyrator as the load for a single 6e5p to drive an 833 at 1500V and 250mA, in heavy A2? I'm working up an 833 amp and was instead thinking of using a CCS on top of the 6e5p followed by an AOT1N60 MOSFET with separate +/- supplies in order to ensure enough drive. Do you think the simple depletion mode mosfet gyrator could do the job? I'm a little doubtful...
The CCS is not an ideal load for an A2 driver. Your load impedance is reduced to low ohms when A2, so you want very load output impedance and high current capability. An IXYS 01N100D is rated 1kV and 100mA. What is the drive required? What is the estimated A2 grid current?
Check the 4-65a SE amp thread where Michael Koster explains extensively the gyrator driver in this topology
Ale
I'll check that out too. I know nothing about gyrators but would like to use one on the output of my phono stage (6e5p again!)
Audio group sounds good, perhaps you could let me know details? Thanks.
I was hoping to try 4p1l two stage amp with 4p1l preamp, it wont leave much gain headroom for quiet recordings but might be worth a go
Enzo
Audio group sounds good, perhaps you could let me know details? Thanks.
I was hoping to try 4p1l two stage amp with 4p1l preamp, it wont leave much gain headroom for quiet recordings but might be worth a go
Enzo
The voltage drive needed will be much less than the 4-65a since the 833 has a mu of 35, but the IXYS 01N100D may be current limited in this application, where I could be in grid current most of the time, hundreds of milliamps of it depending on where I set the operating point. That's why I'm thinking of implementing the CCS on top of the 6e5p, followed by an AOT1N60 mosfet stage (1N60: 700V, 1.3A, only 7.5ohm typical on resistance, low capacitance Crss = 1.8pF). The IXYS 01N100D has an on resistance up to 10X higher than the AOT1N60 and about the same reverse transfer capacitance.
Maybe I could use the AOT1N60 in a gyrator on the 6e5p and save the extra stage...I'll need to investigate further.
The 833 is a devil to drive, unless you go for more than 3000V.
I was never satisfyed with the driver supplying enough current that the "diode for grid" demands.
It seems you have experience with the 833? What types of drivers did you try?
Your description of the drive requirements above is what I was afraid of, and why I'm planning on a separate MOSFET source follower stage with super low impedance and lots of current on demand. Please share your experiences!
Not that much experience with 833A. A few years ago a friend of mine had a collection of NOS 833A and we were building a PP with them. We used 1700V for plate, output trannie was custom built, ...
We tryed all configurations we could think about, using tubes and interstage transformers. Even with impedances as low as 22Ohm, the diode grid effect was visible.
We didnt try mosfet or any SS devices. Hope you can get a better driver than we did.
Hi Enzo,
Gyrators are voltage controlled CCS which emulates the AC response of an inductor in AC without the property of storing energy. There is plenty of info in the forum about using gyrators as loads. Particularly they are very helpful to maintain the bias point despite valve age, as the gyrator will force the anode voltage despite variance in gm/ra. If used in mu-follower mode, the gyrator can complement the driver in an excellent driver for output stages (particularly in A2) as will provide minimum distortion (flat load line in AC) and if well designed all the current needed by the capacitive load (or low impedance load in A2).
Send me a PM re the audio circle details.
Re 4P1L as preamp: the best preamp I've ever built but is a bit tricky on the microphonics. Can result in a fantastic preamp if well implemented though.
Ale
Magz,
Some people have used the 1N60 on gyrators but personally haven't faced that driving requirement myself. I can't see why an 1N60 cannot perform better than a valve in the mu-follower as its gm of 0.9S and 9 ohm of rds will be difficult to match by any valve
A bit of sand is really helpful in this circuit. Just simulate it in SPICE and you will see the results by yourself...Cheers,
Ale
Hi Ale. I've not heard the 4P1L in a preamp but I am running the Pete Millete Low Mu pre amp and I have not heard anyone say anything but "it's the best" and I have a lot of people over....we have a very large audio community here in Michigan.
I feel a good pre amp is a must with a two stage GM70 amp.
http://www.pmillett.com/lowmu_preamp.htm
.
as i remember the microphonics on my 4p1l's were mainly 2 or 3 resonanses around 1500-2000hz i've tried damping the tubes with limited success and am working on mounting all the gain stages on a highish mass rubber decoupled platform.
all this talk of gyrator drivers is tempting me to dig out my sv811-10's my gm70 stereo chassis started out as an A2 push pull 811-10 monoblock!
enzo
all this talk of gyrator drivers is tempting me to dig out my sv811-10's
my gm70 stereo chassis started out as an A2 push pull 811-10 monoblock!enzo
Everything is subjective but if you haven't tried DHT in filament bias (no nasty caps on the signal path that will add that veil to the sound) then you are missing a new world. I've built many preamps using several DHTs such as 01a, 10Y, 30, 30sp, cx112a, 71a, 46, 4P1L and famous 26....
Worth trying, a different experience. Just have a look at the 26 preamp thread and you will find out why there are so many out there using DHT preamps.
Cheers
Ale
Worth trying, a different experience. Just have a look at the 26 preamp thread and you will find out why there are so many out there using DHT preamps.
Cheers
Ale
Hi DJN,
I think its worth considering a line amp that can swing 10-20v pk-pk with a 2 stage design as it could open up the options of driver tubes.
One of the bonuses of diy afterall is not having to conform to industry dogma, and if the amp has a sensativity of 5vrms then you get better snr and less sensativity to hum etc I considered swapping the 6e5p driver and 4p1l line tubes around...
Just remembered your using bi or tri amping so i guess that could make things a little more interesting
enzo
I think its worth considering a line amp that can swing 10-20v pk-pk with a 2 stage design as it could open up the options of driver tubes.
One of the bonuses of diy afterall is not having to conform to industry dogma, and if the amp has a sensativity of 5vrms then you get better snr and less sensativity to hum etc I considered swapping the 6e5p driver and 4p1l line tubes around...
Just remembered your using bi or tri amping so i guess that could make things a little more interesting
enzo
Last edited:
Hi djn,
I would be more inclined for a line stage providing the right impedance and current driving capability to drive your cables and input capacitance. That and its sound are more important than the gain factor. The gain can better be provided by a well design driver in the amp
Cheers, ale
as i remember the microphonics on my 4p1l's were mainly 2 or 3 resonanses around 1500-2000hz i've tried damping the tubes with limited success and am working on mounting all the gain stages on a highish mass rubber decoupled platform.
all this talk of gyrator drivers is tempting me to dig out my sv811-10's
enzo
Hi Enzo,
you're right. The resonance peak is about 1.5kHz. Damping with the classic methods can be very effective but also starving the filaments at about 550mA gives a very good result for reducing the microphonics. The downside is that gm is reduced and rp is increased slightly. Not a big impact to the 1k6 in triode mode.
cheers,
Ale
As I've said before, another way around the cap is to grid bias through a transformer. As a bonus the transformer can step the signal up or down, give complete galvanic isolation (no ground loops), act as a grid choke with a very low DC impedance to ground to stabilize the bias point, and invert the signal polarity to counteract the inversion of the triode. See attachment. Sounds much better to me than grid bias with a cap.
I can't use filament bias because I'm running parafeed with an autoformer, and I return the autoformer ground to the cathode. Since autoformers have no DC tolerance, I can't have DC bias there.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.