Function and form.... / General Chatting / Forums

Forums

4hv.org :: Forums :: General Chatting

« Previous topic | Next topic »

Function and form....

previous 1 2 3 next

Move Thread

LAN_403

Ash Small

Wed Nov 14 2018, 04:33AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Plasma wrote ...

I see, would the voltage and current be in phase, voltage leg would get less affected by a capacitor, thand the lower leg current leg,
Does this sound right?

I don't know... I wasn't sure about this, but I needed an isolated output, and I'd seen this implemented once on a tube amp designed by 'trial and error' rather than theory.

It's customary to fit a capacitor between source and ground (or B- ), in parallel with the source resistor to maximise gain, so I figured maybe connecting the cap to source 'may' have a similar effect.

I can't see it really makes much difference, since the cap is still in series with the load between both points.

I may yet fit a capacitor in parallel with the load, to act as a low pass filter in conjunction with the resistor. I'm still calculating various filter combinations, but it will come down to trial and error (ear). I need high pass, low pass and maybe a conditioning low pass filter. This 'could' go on the output, or on the input of one of the transistors. I'm not even sure I'll need it yet, but it's something I want to experiment with.

EDIT: I've just re-read Plasma's post..... The voltage and current do some strange things, well, mostly the voltage, and Plasma mentioning phase has just caused me to realise why....

there is a point when adjusting the source resistor when there is no voltage across the load, but still maximum current.

either side of this point the voltage rises

I guess at that point they are in phase.

I guess thats where I should adjust it to.

The voltages across the drain resistor and source resistor, I believe, go up and down alternately..... I guess this adds some feedback. It's a remarkable little circuit

EDIT: Actually, the scope is connected to the upper side of the load and ground, which, I think suggests the voltage across the transistor remains constant at all times, or the voltage across the source resistor varies, .....

Shrad

Wed Nov 14 2018, 06:55AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

as you need high pass wouldn't a cascode arrangement be a better fit at higher frequencies?

for filtering, you can also use NFB the japanese way

Ash Small

Wed Nov 14 2018, 07:25AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Shrad wrote ...

as you need high pass wouldn't a cascode arrangement be a better fit at higher frequencies?

for filtering, you can also use NFB the japanese way

I've just had a quick look at cascode, and I don't think it would work with this topology. From what I can see, the cascode circuit requires one side of the load to be grounded, thus not isolated output.

I'm a minimalist, I try to keep my circuits to the absolute minimum component count. I know virtually nothing about feedback, I'm only just beginning to learn about filters. I've worked out that if you have two capacitors connected to the same node, the one with a smaller value will tend to develop a higher voltage across it, all other things being equal, which they are not when you add a couple of resistors...

Shrad

Wed Nov 14 2018, 08:48AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

that's not really an isolated circuit, is it?

maybe an AC coupled differential output would better fit then? just a suggestion of course

EDIT:

like the input stage of this diagram :

2 IRF510adj Dif In

EDIT again:

just found this and thought it was especially on-topic : https://www.edn.com/electronics-blogs/anablog/4307462/Your-friend-the-JFET-

Ash Small

Wed Nov 14 2018, 02:39PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Shrad wrote ...

that's not really an isolated circuit, is it?

maybe an AC coupled differential output would better fit then? just a suggestion of course

EDIT:

like the input stage of this diagram :

2 IRF510adj Dif In

EDIT again:

just found this and thought it was especially on-topic : https://www.edn.com/electronics-blogs/anablog/4307462/Your-friend-the-JFET-

Why is it not isolated output?..... It's de-coupled from the source resistor via the output cap?..... Would a de-coupling cap on the drain make it isolated, or am I missing something?

I'll try googling it.....

Shrad

Wed Nov 14 2018, 03:12PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

when your jfet is conducting your upper output leg is coupled with ground via the transistor (which is the case most of the time) so your output is always coupled to your input in some manner

in the diagram I reference to, both output legs are AC coupled, so isolated

for what I know, a full differential output would be totally decoupled and better achieve your goal... drawback is you have to pair jfets and if you want to parallel them it may be difficult...

Ash Small

Wed Nov 14 2018, 03:56PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Shrad wrote ...

when your jfet is conducting your upper output leg is coupled with ground via the transistor (which is the case most of the time) so your output is always coupled to your input in some manner

in the diagram I reference to, both output legs are AC coupled, so isolated

for what I know, a full differential output would be totally decoupled and better achieve your goal... drawback is you have to pair jfets and if you want to parallel them it may be difficult...

These parallel fine, one of the photo's above has five in parallel, it behaved perfectly... I now see your link is basically a long tailed pair..... I guess I'll just add a second output cap to my previous circuit, it will be isolated both sides then.... Although I do have some iron powder toroids here, and some suitable magnet wire.......

Shrad

Thu Nov 15 2018, 08:08AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

I did my own experimentation with transformer isolation of input and got some promising result by driving a miniature toroid with a PAM8403 class D module, but did not go so far as filtering and actually drive some actual valves

Ash Small

Thu Nov 15 2018, 09:49PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Shrad wrote ...

I did my own experimentation with transformer isolation of input and got some promising result by driving a miniature toroid with a PAM8403 class D module, but did not go so far as filtering and actually drive some actual valves

I've given it some thought, and I'm pretty certain you just need to block any DC component, so one cap should be sufficient, but by then the two I'd ordered yesterday had shown up, I chose 10uF, as they are effectively in parallel, so I'm guessing I only generate half the voltage over either, so they need to be twice the capacitance..... It seemed to make sense....

Anyway, I've now re-drawn the circuit with full isolation. I guess it's an improvement assuming it performs the same.....

1542318558 3414 FT183122 Driver Circuit2

Plasma

Thu Nov 15 2018, 11:22PM

Registered Member #61406 Joined: Thu Jan 05 2017, 11:31PM
Location:
Posts: 268

You're still DC based
About the first,one is a voltage source, ttop leg,the second is a current source.
The one you posted for full isolation will need a cap from top left of the jfet to bottom right,adjust for series,but its still DC,try using 2 MOSFET with diode's at source to clip both ways

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob