• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

parafeed - Gyrator vs CCS anode load

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
dont know until i build it..

It simulates very much like a choke (a big one at that!) but the proof will be in the making and listening.

You can easily improve the PSRR, just stick a v reg infront of it. I am using the salas hv shunt > CCS (soon to be gyrator) > parafeed edcor.
 
Wait - This will limit you to 0.7v on the gates due to the forward voltage....

Edit: Ignore me, back to back is not anti-parallel :crazy:

The DN2540 is rated for +/- 20V gate-source but at normal operating temperatures saturates at about 0Vgs, so a single 15V zener connected with it's anode to the DN2540 gate (as you would a P-channel) works fine for protection.

PS Another option for driving headphone or speaker loads is this balanced current totem pole push-pull circuit using a gyrator. The zener is shown as well as a nice cascode.

Another big difference between the Gyrator/CCS and choke or series feed OPT is the need for higher B+ and consequent dissipation in the active load.
 

Attachments

  • CascodeGyratorAT.JPG
    CascodeGyratorAT.JPG
    38.5 KB · Views: 987
Last edited:

iko

Ex-Moderator
Joined 2008
Another idea for the gyrator, which I borrowed from Anatoliy. I have built this headphone amp a few days ago and have been listening to it; I think the sound is very good, very dynamic, smooth, no lack of bass, nice full mids. The autoformers I got from a surplus shop for a few dollars. These are the type of line matching transformers used in store ceiling speakers.

I should add that I personally prefer using the mje350, since it's easily available at local shops and can get them easily. I doubt it'd make a difference in sound, when compared to using a dn2540. But that is just a guess.
 

Attachments

  • amp-hp-6s45pe-autof-2a.jpg
    amp-hp-6s45pe-autof-2a.jpg
    45.8 KB · Views: 839
  • amp-hp-6s45pe-autof-2a-gain.jpg
    amp-hp-6s45pe-autof-2a-gain.jpg
    64.3 KB · Views: 819
Looks fine to me. Since your gate is driven by the plate supply, I think protection diodes aren't necessary. A short from output to ground would be a problem, but that should be easy to avoid.

BTW, with the cascode, a low voltage JFET is an option for the bottom device.

Sheldon
 
Another idea for the gyrator, which I borrowed from Anatoliy. I have built this headphone amp a few days ago and have been listening to it; I think the sound is very good, very dynamic, smooth, no lack of bass, nice full mids. The autoformers I got from a surplus shop for a few dollars. These are the type of line matching transformers used in store ceiling speakers.

I am just this minute drawing up a similar version (but with much smaller cap and a pnp cascode). I'll sim it to make sure no boo boos.
 

iko

Ex-Moderator
Joined 2008
As you say, a cascode gyrator might be even better. This one simulates better: green line is cascode, blue line the one I built. It's easy enough that I might give it a try tonight.
 

Attachments

  • gyrator-casc-1a.png
    gyrator-casc-1a.png
    8 KB · Views: 985
  • gyrator-casc-1a-comparison.png
    gyrator-casc-1a-comparison.png
    8.5 KB · Views: 951
With the D3a and a fixed cathode voltage, I'd recommend looking at a CCS version also. With both cathode and plate voltage fixed, the plate current seems to drift around with the D3a. With a CCS the plate voltage only moves a few volts. To me the CCS seems to result in a more consistent op point. Also there's no capacitor to charge and discharge with varying signal levels, were there to be any 2nd harmonic distortion in the circuit.

Some say the gyrator sounds better than the CCS in some circuits, but I'm not getting that impression at all, over a few different amplifiers where I used both and particularly in some recent circuits using the 6C45Pi and the D3a in triode mode. My comparisons are based on using the "mu output" or the center tap, however.

The big difference to me in the gyrator circuit (MOSFET style) is that the resistance between plate and source can be made larger, thus reducing the influence of the MOSFET gfs characteristic. However, the DN2540 has a relatively linear characteristic. I find that the resistance I need to get 10-20 mA current with a DN2540 in CCS mode is enough to get excellent circuit performance.

I agree with Sheldon as long as the capacitor doesn't get charged above the max Vgs there should be little risk. Still, I only omit the diode in the CCS style circuits... I've not yet blown up a MOSFET (in an audio circuit that is...)
 

Attachments

  • CCSversion.JPG
    CCSversion.JPG
    28.9 KB · Views: 971
test bed is complete... just got to get round to pulling the guts out my headamp and sticking this in there

An externally hosted image should be here but it was not working when we last tested it.


note - the lead length is a bit long on the dn2540s, hopefully i will get away with it but if they oscillate i will whip them out and drill the heatsinks to allow lower mounting.
 
Last edited:
Hi All

I have a cathode follower and I want to try a gyrator there, i have test it with a N mosfet in mirror drawn circuit,, I am happy to known how the calculations go about this circuit, I want to test it some time in a hybride, the cathode follower has a idle of 20 mA.

Sorry but I have never try a gyrator, i have a CCS loaded one playing in my room.

regards.
 
Hi

I have made a example what i do, it is a dc coupled hybride.

now I use a CCS for the first and second tube, the sound is hiss to much, the mid is to sharp, feeling of compression of music but this can be the mp3 I did listening to, but resolution of music is exellent, maybe mp3 is much irritant when listening on high end.

foto 1 is a test and the amp I now listen to is foto 3, 2 is the test measurement in multisim of the gyrator in work incl one for the cathodefollower..

I couple the cathode follower dc to the mosfets with a offset ic control it.

I have change the tube EL84 for 6n6p this tube looks more harsing, sharp mid, with the el84 it is was warmer, maybe because of the double triode being paralell.

How do you calculate the gyrator? for example a 1 mA tube ecc803.

thanks in advance.
 

Attachments

  • gyrator test.jpg
    gyrator test.jpg
    179.6 KB · Views: 555
  • ac-analyse.jpg
    ac-analyse.jpg
    54.6 KB · Views: 360
  • working hybride.jpg
    working hybride.jpg
    81.9 KB · Views: 492
Last edited:
Hi

I have made a example what i do, it is a dc coupled hybride.

now I use a CCS for the first and second tube, the sound is hiss to much, the mid is to sharp, feeling of compression of music but this can be the mp3 I did listening to, but resolution of music is exellent, maybe mp3 is much irritant when listening on high end.

A properly functioning gyrator doesn't add significant hiss or cause listening fatigue, but those are classic symptoms of RF oscillation.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.