Electronic Choke for Valve Amplifiers

[psaitz]

Hi!

Menno van der Veen - located in the Netherlands - offers an electronic choke. The description an technical data sounds great (in theory).

Has anyone tested/used this electronic choke in a valve power amplifier? Is it really better than a classic choke? I'm interested in the schematics too. Where can I get it?

Thanks for your help
Peter from Vienna/Austria

 

[TheGimp]

It looks like a gyrator.

Do a search on "Gyrator Plate Load" and you will find lots of discussion about it.

It will not store energy like a real choke. Otherwise it will work like one in a power supply for ripple rejection.

As a plate load it will not allow the full voltage swing of a true inductor plate load.

 

[Wavebourn]

Gyrator plate load and gyrator instead of a power filter choke are very different beasts. The first is a good improvement, the second is suboptimal. Beauty of PS filter choke is, it stores an energy. High impedance in frequency band of amplified signal is it's minus. If to use active devices to filter PS ripples, I would rather prefer such topologies that offer low output dynamic resistance, instead of imitation of choke weak points.

 

[LinuksGuru]

Personally I use 4H/400mA fixed air gap toroidal chokes. Quite small and efficient.
Additionally, I have built regulated stabilized power supply 500V/0.5A.
IMHO optimal solution is to use traditional CLC or stabilized regulated power supply, or even better, combination of both.

 

[Bigun]

Another thing to bear in mind, real chokes create magnetic fields around them (leakage flux) which can couple hum into your amplifier despite good B+ filtering.

Nevertheless, I like real chokes for their simplicity (ease of design, construction, debugging and maintenance) - I suspect they might be more reliable but I've no data to support that.

 

[ArtG]

Personally I use 4H/400mA fixed air gap toroidal chokes. Quite small and efficient.
Additionally, I have built regulated stabilized power supply 500V/0.5A.
IMHO optimal solution is to use traditional CLC or stabilized regulated power supply, or even better, combination of both.

Can you suggest a source for a small quantity purchase of such an item?

 

[LinuksGuru]

Can you suggest a source for a small quantity purchase of such an item?

I have ordered bunch of them straight from the factory along with other items for our business, I did not found anything comparable in any large European electronic store.
If you are in Europe, I can sell them for 25 GBP/pc, but if you are outside Europe, shipping cost will be unreasonable high and prohibitive (choke weights 1.8 kg).

 

[rsumperl]

Personally I use 4H/400mA fixed air gap toroidal chokes. Quite small and efficient.
Additionally, I have built regulated stabilized power supply 500V/0.5A.
IMHO optimal solution is to use traditional CLC or stabilized regulated power supply, or even better, combination of both.

Can you post a schematic of your 500v supply?

Thanks
Ray

 

[DF96]

I am intrigued by Fig 4 in the link from post #1. It shows the decay with a 10H filter choke in the PSU, between two 330uF caps. There are two prominent components which shift up in frequency by a factor of about 3! I would have expected LC ringing to be rather more stable in frequency. Of course, we don't know exactly what the initial stimulus was, and whether DC through the choke is changing its inductance.

Any ideas?

 

[LinuksGuru]

Can you post a schematic of your 500v supply?
Ray

I'll do it when finish everything. Right now I have working stabilizer built on the teflon board very rough way. The next things to do are turn on delay (40 - 60 sec to avoid cathode poisoning), and PCB. I do not wish to post schematic for device which is not tested enough.

 

[Miniwatt]

I compared the 'E-choke' in my 6072 preamp with 8uF-10H-10uF-20H-20uF PS. The new PS was 8uF -E-choke-330uF.
This was a major difference. Though the conventional setup was extensive, this made me realise the importance of clean power. It was like a different and definately better preamp. I'm very happy with the result, I haven't yet tried it in power amps though.
I asked Menno what value in Henries it would represent in simulations and the answer was 70! That's a big a** choke!

I'm not sure which choke I prefer at this moment, I take the 'energy storage' aspect of a conventional choke very serious.
Okay, I just became curious, there will be some testing in the future..

 

[[email protected]]

I’m not sure what in the name gyrator plate load and gyrator. What are the differences?

I got two monster caps from a friend of mine for free, 4400 microF/500 V (EPCOS). If the good thing about a choke is that it stores energy my caps must be good in my coming tube power amplifier. They store energy pretty good.

My idea is to have some regulated stabilized power supply, maybe something like the electronic choke, for the driver circuit and then my monster caps maybe with a choke for the output stage. Does this sound like a good way to use my caps?

 

[Miniwatt]

What size are those caps? A 2 liter can?!
My present was a bit smaller, two 680uF 500V. BC components, very nice, must cost quite a bit, but I'm hesitant to use them, afraid they might be too slow, and -if there's such a thing- too large value. But I haven't tried them though.

 

[[email protected]]

More like 1 liter

Combined with smaller caps it might not be slow.

 

[ArtG]

They are rather large, I imagine! I have in my hand, at the moment, a 1400uF @450VDC. It measures 2.5 in. wide X 5.75 in. long. It's rather fast, actually-it's a "Computer Grade" cap. I hope some hefty bleeder resistors are in place with those caps.
Q/ What kind of a computer uses a 450V cap?
A/ The kind you don't want to own!

 

[gingertube]

Its been discussed.
See here:
http://www.diyaudio.com/forums/power-supplies/155657-mosfet-gyrator-using-irf840.html
Cheers,
Ian

 

[[email protected]]

They are rather large, I imagine! I have in my hand, at the moment, a 1400uF @450VDC. It measures 2.5 in. wide X 5.75 in. long. It's rather fast, actually-it's a "Computer Grade" cap. I hope some hefty bleeder resistors are in place with those caps.
Q/ What kind of a computer uses a 450V cap?
A/ The kind you don't want to own!

It is normally used in a UPS for computeres.

 

[opcom]

I can see that the main use for the E-choke is not to replace a choke but to obtain the silencing benefits of one.

I will say that there are applications where the choke offers benefits of voltage regulation and improved efficiency of the power supply as well as better power factor. Rectifier peak currents are much less a concern when a stout choke is used.

Very high power >1KVA amplifiers I always use a choke input, and sometimes to reduce transients accompanying class AB operations, will use a resonant choke such as 2H in parallel with 0.88uF is used, before the capacitive filter.

I prefer the nasty old chokes because the voltage regulation is much better than a capacitor only. This is not very important for class A amps, or even for those that enter AB only near their limits with max/mincurrent variation of <1.5:1, but is very important when the current varies widely as amp designs approach class B end of things, varying more like 2:1 to 4:1.

I am running a 4800V 2A supply, choke input followed by a 31uF oil capacitor. At this time in a 6KW class AB amplifier under construction. The power supply is complete and tested. What I can say is that with either the 30H choke or the 2H/0.88uF resonant choke setup, the ripple is 1-2% and regulation about 1%, that is fantastic. On step loads such as 300mA-2000mA, which is what you might get in an AB amp especially for RF, or for bursts of sound (1812 overture cannon anyone?), the resonant choke is a must. The type of amp matters less than the profile of the load on the power supply.

About amps - note the 6L6 and EL34 conditions, since it's common:

push pull class A:
120-140mA current ratio 1:1.16
134-155mA current ratio 1:1.16

push pull class AB1:
88-132mA current ratio 1:1.5
88-140mA current ratio 1:1.6
116-210mA current ratio 1:1.8

push pull class AB2:
78-142mA current ratio 1:1.8
88-205mA current ratio 1:2.3

push pull class "B":
50-182mA current ratio 1:3.6

I believe it would be fine to use Mr. Vanderveen's electronic choke within its ratings in any amp that has already a choke, and that similarly pleasant improvements will be made. I think it ought go after the choke in those cases but someone who knows more about it should comment.