'E'-Choke

If they glow, this is the place to be
Max N
Old Hand
Posts: 1453
Joined: Thu Aug 16, 2007 6:10 pm

#136

Post by Max N »

Hi Mike
I have built up one of the boards and installed it in the amp.
It seems to be working OK but the heatsink is getting very hot.
Voltage drop across the module is about 45V, and the load is about 130mA.
Configuration is:
Rectifier > choke > 100uF > e-choke > 1000uF > load
Voltage at C1 is about 348V
Voltage at C2 (load) is about 303V
Ripple at C1 is about 500mV RMS
Ripple at C2 is about 1mV RMS
CG is 47uF

Does this seem as if the module is working OK? The voltage drop seems high to me?

Thanks

Max
:)
Max N
Old Hand
Posts: 1453
Joined: Thu Aug 16, 2007 6:10 pm

#137

Post by Max N »

Mike, do you have any more kits available for sale? Preferably with the resistor modification?
Cheers
Max
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#138

Post by Mike H »

Yes. I've increased the 'available' number on the web page to 10.


Max N wrote:I have built up one of the boards and installed it in the amp.
It seems to be working OK but the heatsink is getting very hot.
Voltage drop across the module is about 45V, and the load is about 130mA.

Does this seem as if the module is working OK? The voltage drop seems high to me?
Yes it is, should be more like 27V. :!:

Try taking the 47uF off CG, as it might be leaky. The current sink tolerance is very low at this point, as it goes through a 470k resistor, so ANY leakage will have a large effect. 1uA will drop it half a Volt, so sounds like it's dropping about 20V (40uA :?: ). And the Voltage here is about 4 - 5V above OUT. I'm assuming you've still got the 10uF on the board?

I'm also thinking 1,000uF on the output is very large, 100uF would be plenty I would have thought.

@ 130mA the MOSFET should be dissipating >=3 Watts (25V drop).


 
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Max N
Old Hand
Posts: 1453
Joined: Thu Aug 16, 2007 6:10 pm

#139

Post by Max N »

The 47uF is on the board in the C1 position. Its a radial electrolytic, a cheap Jamicon from Maplins.
I'll replace it with the 630V 1uF that you supplied in the kit.
I'll measure some voltages and report back
Cheers
Max
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#140

Post by Mike H »

Ah right, yes try that. Image
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8882
Joined: Mon May 21, 2007 9:42 pm

#141

Post by Paul Barker »

Mike H wrote:Yes. I've increased the 'available' number on the web page to 10.
Mike I can't see E choke anywhere on Living in the Past.

Either you have taken it down or you need to take a look at how your website is designed.

Also when you use your search facility it just routes you to a standard internet search so, a chocolate fire guard for searching your site.

Not so much a criticism as an observation which may help you.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#142

Post by Mike H »

A. This particular 'offer' was meant to be exclusively for Audio Talk members only as it's at cost price, hence you can't find it through the web site. It was only meant to help somewhat with the initial cost of having 50 boards made so the number available always was limited and not expected to last very long. Ultimately the idea is that the web site will show ready-built modules which will have a profit mark-up accordingly. Only haven't found time to do it yet.

B. The link to that particular page is in the Buying And Selling > E-Choke V4 Boards topic, I've just repeated the link post again to be sure, as there's a few more available now (I moved the goalposts).

C. I'm not clever enough to write search engine code, although I tried something like it once and it wasn't very good, and anyway there's little point since Google works so well.

Oh except, the page has also got <meta name="robots" content="noindex"> so Google won't know about it either.

I.e, it's supposed to be a secret just between ourselves.

Image


Next?
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Max N
Old Hand
Posts: 1453
Joined: Thu Aug 16, 2007 6:10 pm

#143

Post by Max N »

This is a re-post from my 300B PSE thread in 'Everyone's Projects', just so that the E-choke info is all in one place:

Following Mike's advice I removed the 47uF C1 and just fitted the 1uF that Mike supplies in the kit. This has done the trick, voltage drop is now 27V (which I reckon is near enough what we expect?) and I can keep my finger on the heatsink (just)
B+ ripple is now about 2.5mV RMS

While I had the amp on the bench, I took some measurements. The amp is flat from 10Hz to 90kHz, there is no hump in the bass.

Most importantly, it now sounds very good to my ears. I just have to get the other one running again and then I can listen in stereo. :D

Thanks again Mike :D
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#144

Post by Mike H »

Has it still got the fuse on it?


Assuming you are Matthew (lol) hopefully posting you the further 4 modified sets on Monday.

Edit: yes 27V sounds more like it.

In your case the MOSFET is dissipating about 3.25W so will be a bit hot yes.

 
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Max N
Old Hand
Posts: 1453
Joined: Thu Aug 16, 2007 6:10 pm

#145

Post by Max N »

Mike H wrote:Has it still got the fuse on it?


Assuming you are Matthew (lol) hopefully posting you the further 4 modified sets on Monday.

Edit: yes 27V sounds more like it.

In your case the MOSFET is dissipating about 3.25W so will be a bit hot yes.

No, the fuse has been replaced by the 10R - I think its all as per your latest spec.

Yes I am Matthew :D
User avatar
Nick
Site Admin
Posts: 15711
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#146

Post by Nick »

No, I am Matthew !!!
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Cressy Snr
Amstrad Tower of Power
Posts: 10552
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#147

Post by Cressy Snr »

Well I'm Spartacus!
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#148

Post by Mike H »

Clear off ... Image


:lol:

Max N wrote:No, the fuse has been replaced by the 10R - I think its all as per your latest spec.
Okey-dokey then! Image


 
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Max N
Old Hand
Posts: 1453
Joined: Thu Aug 16, 2007 6:10 pm

#149

Post by Max N »

I managed to resurrect the fetreg that I blew up a few days ago - had to replace the mosfet and R5, which had gone open circuit. Its now working fine, so I have two working amps.
So to re-cap:

With the original fetreg specification I had one amp working OK and one buzzing.

I then fitted the latest spec, with the SMD resistor mod, and with a 47uF in the C1 position. This worked but the voltage drop was higher than expected. Mike diagnosed this as a leaky C1.

I then fitted the original 1uF 630V cap (as supplied in the kit) in the C1 position. The voltage drop now is 27V as expected, and the heatsink is hot but not too hot.

I have 100uF as the first cap and 1000uF as the second cap. Ripple is less than 2mV RMS.

So to summarise, the latest spec of fetreg with the smd resistor modification is stable even with very high values of input and output cap. Ripple attenuation is very good even with just 1uF in the C1 position.

Am I right in thinking that with the 10R in series on the output, the output impedance is quite high? In any case, that is one reason why I have stuck with a very large output cap - to ensure a low impedance supply down to low frequencies.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#150

Post by Mike H »

No that's one of Nick's tricks, it isolates the output capacitor from the regulator output which makes your amp sound betterer. He can explain it better than I can at the mo ('cause I've forgotten the intricacies as usual).

I wouldn't have though 10R would make it that much more "high impedance", you could try 100uF and see if it makes any real difference really.

Otherwise all sounds good! Image

You could still increase CG if you can find a less leaky electrolytic? So far I'm using 10uF Panasonic radials they seem OK.


 
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
Post Reply