Inductor Loading Adventures

Status
Not open for further replies.
Jump to Latest
#63
davidallancole said:
flg, you don't happen to have a image of what your square wave is looking like at 1kHz, do you?
Click to expand...
Sure, it was a page back... But it looks like this... Actually this pic would represent the performance at about 100Hz. @ 1kHz the "tilt" is barely noticable...
 

Attachments

  • sej tilt_thumb.png
    sej tilt_thumb.png
    1.4 KB · Views: 250
#65
Last night I got in a few more minutes of experimenting. If I remember right, I tried a load R of 15 ohms, the gain went up substantially and so did the Sq wave distortion at the same freq. I unloaded the output totally and saw a gain of about 22! Hmmm :smash:
BTW, the amp did not seem to like no load with a square wave. The waveform got real distorted.
Also when sweeping the frequency up with a sine wave, I got to about 3200Hz and the inductor gave off a nice sounding ringing noise, although the scope showed nothing unussual? Resonance 😀
 
#67
flg said:
Also when sweeping the frequency up with a sine wave, I got to about 3200Hz and the inductor gave off a nice sounding ringing noise, although the scope showed nothing unusual? Resonance 😀
Click to expand...

I remember getting a ringing noise from my air-core inductor once--seems to me I was driving it very hard. I wish I had kept notes. I believe that turning down the preamp volume reduced the ringing in my ZV7-T.

edit: the more I think about it... I believe that one side of the balanced power fet stage had a weak connection; this made it more or less single ended. I did discover and correct the bad connection and believe I solved the problem.
 
#68
So, where are we? Well, tonight I got a few minutes so I decided to try something. I've been thinking about it on and off all day. Hope my boss dosen't read this :cannotbe: I'ld really like to explain why but, the abillity to think in a real explanitory terms isn't with me right now... And I'm not sure there really is a good explanation...
I'm thinking this inductor load is causing a very frequency dependant load. Mostly due to the AC (read Xl) load increasing as frequency goes up. I still don't understand why I simply don't see gain going up with freq, but, the square wave distortion, is telling me that some freqs have higher gain than others. The sine wave sweep does not show anything with significance to claim this but no one else has a better explanation do they?
It would be expected that the inductor would reach a point in it's Xl where the transconductance of the JFET would just run out and that's all the gain I'm going to get. I think that would be around A=30. That figure would be lowered by the source resistor and the 7.5 ohm load too but, whatever😕. I'm thinking that up beyond 1K or so, where the sq wave distortion goes away, I'm at that point. The Xl would be 100 times the load! And last night I saw an Open Loop, Un-Loaded gain of 22. O.K. not bad...
But, if I could stabilize the gain at some constant, like with a resistor, instead of the constantly rising Xl of the inductor, I would not see as much sq wave distortion. So I put a 55 ohm R in paralell with the L...
Very little change??? So I put a 27.5 ohm in paralell with the inductor and still only slightly better??? I guess I can still paralell my 2 big 27.5 ohm Rs and check it again but that will just be killing the gain somore 😕 So I did that and it's Tilt is 1/2 the slope it was when I started... but the gain is back down to about 8.5 :cannotbe:
Does anyone understand what I'm trying to do here??? Am I out in space or what???
 
#70
Magura said:
Does it change anything if you feed the amp input with less than the max voltage you previously tested it with? ..."🙂
Click to expand...
Nope 😀 Like 100mV in? Nota, same 'ol, same 'ol... Max voltage was originally 22-24 Volts, but I test at 10V out and 2-4 V out too :smash:
Very frusrating, I want to make this work 😕 Taking a break tomorrow. Going 250 miles south to see some Drag Racing! You have that where you are don't you? I'm not a 1 ball sport fan... Gotta get up by 4:00am Later 😀
 
#72
Magura said:
This is getting plain odd!
The only reason left for your amp not to work, has to be the inductors :whazzat:
Do you have an alternative to the inductors you use now? I'm thinking big trafo or the like.
Magura 🙂
Click to expand...


Yes, would answer that question. But, as has been suggested, and I have proven to be true, everything I've tried was not designed to sustain DC current and maintain the mesasured inductance, without saturation, except for these hammond units. They actually make a 100mH 10A choke but it is $$$$$. That's $250 USD ea. :bigeyes:
Let me recap those experiments a little for the other 8 DIYer's. I have a dual primary 800VA toriod. I believe these types of deals are actually laminated also with a long strip of very thin "flat stock" core material, wound up to look like a donut. I measured this unit at 275mH/winding and well over 1H in series 😀 When tested with 1 winding it performed quite good but exhibited the dreaded curvature of saturation charicteristics in the Tilt area of the sq wave at low freqs. Yes, it had the tilt also. The saturation of sq waves started at a higher freq though, and the fl -3db point of sine waves probably suffered also at about 100Hz? I believe due to saturation, I was not enjoying all 275mH. I tested it with a series connection and it showed better freq response but more saturation :bawling:
I also tested this EI core monster 15-20lb (40k) thing my Dad had. He had 2. One of them was in a 14V 50A regulated supply. I estimate it was about a 1kVa unit. I did not measure the inductance but it was .2 Ohm DCR. It seemed to do better in the lower freqs and in the saturation departments, than the toriod. It did not exhibit quite as good high freq performance. I beleive the core material and thickness of the laminations were the heavy contributors there...
So, Other inductors appeared limited due to the saturation problems. They were all designed for AC input at 60Hz only and another freq, espeacially lower, is not optimum. They may have had more inductance in the AC realm but I possibly could not take advantage of it due to the Amp's DC Iq :whazzat:

Now for some real DIY 😀 I have removed the screws holding my Hammond 195T5 together to find it is indeed an air gapped unit :bigeyes: Acttually, 4 pieces of mylar looking stuff measuring about .0095"-.0105" in thicknes, totalling about an .040" gap. For all intents and purposes, air, to the rest of the core. So, I have a little experience, obviously, with this science but I'm no expert. I've fabricated little tiny flyback transformers for SMP's that I ussually used about a .007" gap to increase the inductance about 3x-10x if I remember correctly 😕 Wish I had my notes :xeye: This would lower the saturation current though if I also remember correctly. Does anyone have any guestimations for me? I'm guessing doubleing the gap to about .075"-.08" might be a good starting point???
:smash: :smash: :smash:
 
#74
How much power do I "need" from this amp? You must have wanted to say, How much do I want? 😀
My original expected target was about 25 WRMS :idea: I could settle for 15W but the abilty to do 35W would be extremely satisfying...
Having said that, I just want to know what it takes to acheive this type of topoligy with good linearity :smash: If it's more "doable" with a midbass-mid freq and higher bandwith, so be-it. The lower end of the desired wattage would be useful for that purpose. Since I've also decided to try to make the Power JFET be the gain device, I beleive 1.5Amps or so is an operating area to keep in mind. maybe using 2 of them as I originally was working on...
 
#75
Naah, I did actually try to use the word "need", though I know that actual need should be of no importance 😀

If you go down in wattage, you could just try with a spool of wire, and consider such an air core inductor. As I recall I used a spool of 18avg hook-up wire for my initial tests.


Magura 🙂
 
#78
The sine waves look good. I haven't done actuall THD measurements yet though...
And, Impregnation with something might be good after hearing the inductor ringing up around 3k. I also started dissassembling one to play with the gap shims. It rings at other freqs without being tightly clamped together... Impregnating is something to thing about in the final build I would think...
So far I tested it with .065" air gap (.040 original + .025 more). I saw a very slight improvement...
:smash: :smash: :smash:
 
#79
i like this thread.

hey this is a cool thread.

Magura:

do you have technical info about those 80mH air core chokes you wound? wire gauge and what type of former you wound it on, turn number etc??

also, could this design be done with batteries?

any THD measurements?
 
#80
Trust me you don't want a battery supply for a zen amp. I went there with a 5 watt SOZ and while it sounded lovely it always
seemed that just when I wanted to listen a little longer the batterys went dead. If you do I subjest having a battery raited for
at least 20 times your current draw. That is a 40a hour battery
for a 2a current draw.
 
Status
Not open for further replies.
Top Bottom