K&K Audio / Lundahl Transformers

Welcome! Need support, you got it. Or share youe ideas and experiences.

Return to K&K Audio / Lundahl Transformers

K&K Audio / Lundahl Transformers

Problems with LL1660 driving 6c33! Need help!

217.236.60.111

[ Follow Ups ] [ Post Followup ] [ Thread: Display Email Next ] [ K&K Audio / Lundahl Transformers ]

Hello Kevin,
i have problems with my LL1660/20ma interstage transformers to drive directly the 6c33 triode.
I get a freq. cutoff 1-2 db continously from 10Khz. The -3dB point is 23Khz. I use a 6c45 with 40mA in ALT S 1:1,25 connection (gain ~50). I also tried the 5842 with 20 mA in that connection. On the scop you see a cutoff 1-2 dB continously from 10Khz. Probably it happen because the input capacitance of the 6c33 is ~11pF. But what could i do? I also need a gain in this circuit near 50 or more!
Thanx a lot
Hauke

Follow Ups:

Re: Problems with LL1660 driving 6c33! Need help! - vry 02:49:33 10/29/03 (5)
Hi
A few possibilities:
1) May be worth double checking the connections on the interstage, particularly to ensure that the secondary is connected the correct way round (ie pin 16 to ground and pin 21 to grid, and not the reverse), likewise for the primary. May make some difference.
2) Try decreasing the current - from the datasheet, looks like you may be approaching saturation? As this lowers the permeablility of the core, the leakage inductance increases which would certainly reduce top end bandwidth.
3) The input capacitance of the 6c33 is roughly
[(gain+1) x Cga] + Cgk
which I make to be about 120 pF (if you trust the published data! It may well be higher)
You then need to multiply this by the sqaure of the transformer stepup ratio, 1.25*1.25 = 1.5625. Call the end result 200pF to include strays. This is about 40k at 20kHz, well above the Ri of the driver, so I doubt it's that.
An easy way to check is to parallel a 1k resistor with the transformer primary. This will pretty much halve the output resistance of the 6C45 or 417, taking it nearer to 500 ohms - see if that gives you an improvement in bandwidth.
Hope this is of use.
Best of luck!
- Re: Problems with LL1660 driving 6c33! Need help! - Haukito 08:14:34 10/29/03 (4)
  Hello,
  i checked again the hole circuit. The LL1660/18mA 1:1,25 was connected right. At pin 16 i connected the bias for the 6c33. I now lowered the gridstopper of the 6c45 to 150 ohm. I removed the 6c33 and grounded the LL1660 on pin 16. I paralled a 10KOhm resistor to the output of the LL1660. I measured 24KHz -3dB upper frequency limit, without (minimal) cutoff . Also i tried different Ia for the 6c45 15-22mA. Whitout Load on the 6c45-LL1660 circuit i measured upper frequency limit up to 28KHz without cutoff.While i measured a cutoff on the grid of the 6c33 begins from ~10KHz. The total circuit shows a cut off under max. Load 21,25Vpp into 8 Ohm Load (~ round about 7 Watt output power):
  14Hz 70% (-3dB)
  1KHz odB
  15KHz 77,5%
  20,5KHz 70% (-3dB)
  What could i do? Seems that the driver circuit including LL1660 doesn't like the heavy capacitivy load from the 6c33. I think this is possibly the 80% problem the other 20% causes the output transformer of the 6c33.
  I think i have to build a new circuit with a input tube (gain ~ 20) and a stable output driver stage with low Output-R, while the current consumption is limited by the LL1660/18mA 1:1,25 to ~20mA for this tube.
  Any ideas? Or could i drive the 6c33 with one tube and the LL1660?
  Many Questions?!?!
  Thanks a lot!
  Haukito
  - Re: Problems with LL1660 driving 6c33! Need help! - vry 03:20:10 10/30/03 (2)
    Allo Haukito,
    Let's just ensure I'm reading you right. You're saying the measurements for the driver stage alone are:
    1) With driver stage unloaded, -3db 28kHz
    2) Driver stage loaded with 10k on secondary, -3db 24kHz
    Please can you just clarify what the driver stage -3db point is with the 6C33 connected?
    Might be tricky if the thing oscillates with the probe attached...
    [does that sound filthy to anyone?]
    As for the paralleled ECC88/6922 suggestion, I think that combination will have an even higher Ri than the 6c45 (from memory), so an E55L may well be a better bet.
    cheers
    - Re: Problems with LL1660 driving 6c33! Need help! - Haukito 04:07:30 10/30/03 (1)
      Hello VRY,
      The total circuit (with 6C33) shows a cut off under max. Load 21,25Vpp into 8 Ohm Load (~ round about 7 Watt output power):
      14Hz 70% (-3dB)
      1KHz odB
      15KHz 77,5%
      20,5KHz 70% (-3dB)
      The -3dB at 20,5 KHz is not the problem. The problem is that the amplitude begins to drop off at 10KHz.
      Greetings Hauke
      - I think it's the OPT... - vry 07:07:58 10/30/03 (0)
        
        In Reply to: Re: Problems with LL1660 driving 6c33! Need help! posted by Haukito on October 30, 2003 at 04:07:30:
        
        Hi
        The -3dB at 20,5 KHz is not the problem. The problem is that the amplitude begins to drop off at 10KHz.
        Well, if you have a first order filter that is -3db at 20kHz, it will be down about -1db at 10kHz, so you should expect to see it roll off from around 10k! A rough calculation for 15kHz and I get 0.8 - not far from the 77.5% you saw.
        So, I suspect much of the rolloff is coming from the OPT - which one are you using, and what's the primary current you are running.
  - Re: Problems with LL1660 driving 6c33! Need help! - Haukito 08:39:22 10/29/03 (0)
    Hello,
    what about 1*6922/ECC88 with 2 triode system parallel. Each system with ca. 10mA and connect to the LL1660/18mA 1:1,25.
    Would this be better than 1* 6c45 with ~20mA to drive the capacitivly fuc.... 6c33?
    Greetings Haukito
Re: Problems with LL1660 driving 6c33! Need help! - vry 02:49:29 10/29/03 (1)
Hi
A few possibilities:
1) May be worth double checking the connections on the interstage, particularly to ensure that the secondary is connected the correct way round (ie pin 16 to ground and pin 21 to grid, and not the reverse), likewise for the primary. May make some difference.
2) Try decreasing the current - from the datasheet, looks like you may be approaching saturation? As this lowers the permeablility of the core, the leakage inductance increases which would certainly reduce top end bandwidth.
3) The input capacitance of the 6c33 is roughly
[(gain+1) x Cga] + Cgk
which I make to be about 120 pF (if you trust the published data! It may well be higher)
You then need to multiply this by the sqaure of the transformer stepup ratio, 1.25*1.25 = 1.5625. Call the end result 200pF to include strays. This is about 40k at 20kHz, well above the Ri of the driver, so I doubt it's that.
An easy way to check is to parallel a 1k resistor with the transformer primary. This will pretty much halve the output resistance of the 6C45 or 417, taking it nearer to 500 ohms - see if that gives you an improvement in bandwidth.
Hope this is of use.
Best of luck!
- Please excuse the double post; blasted browser...[nt] - vry 02:51:44 10/29/03 (0)
Re: Problems with LL1660 driving 6c33! Need help! - KevinC 06:54:49 10/27/03 (5)
It appears that your roll-off may have to do with three (and perhaps more)parameters. The input capacitance of the 6C33 may be interacting with the source impedance (6C45 plus transformer ratio plus grid stopper)to roll off the high frequencies. You can lower the source impedance and get the same gain by switching to a triode connected E55L (8223) input tube and/or lowering your grid stopper resistor value. This tube has a substantially lower plate impedance and, in my limited experience, sounds better than a 6C45. Alt S is a good connection alternative to use.

Kevin Carter
K&K Audio
www.kandkaudio.com
- How to connect the Grid 2 and 3 for the E55L when using as a triode ? - akltam 18:51:41 10/28/03 (1)
  Thanks.
  Alan
  - Re: How to connect the Grid 2 and 3 for the E55L when using as a triode ? - KevinC 19:06:10 10/28/03 (0)
    As I recall (which is a chancey business), I connected them both to the cathode. Of course, it is more usual to connect grid 2 to the plate, but I didn't try that. An audio friend whose ears I trust said that it sounded better connected to the cathode.
    
    Kevin Carter
    K&K Audio
    www.kandkaudio.com
- Re: Problems with LL1660 driving 6c33! Need help! - MaciekP 10:11:30 10/27/03 (2)
  hello,
  Besed on my experiences, E55L triode connected sounded very similar to e810f triode connected -DARK. I think, this is due to high internal capacitance. What about trying to use your IT as a plate choke and use good capacitor?
  Yes, you will not get any step-up...but frequency response should be fine.
  Maciej
  - Funny, but the 6C45 sounds bright to me... - KevinC 11:39:29 10/27/03 (0)
    It's all about system balance. The 6C45 doesn't compliment the rest of my system components, but the 8233 (E55L) does. There ain't no right or wrong, just what you like!
    
    Kevin Carter
    K&K Audio
    www.kandkaudio.com
  - Re: Problems with LL1660 driving 6c33! Need help! - Haukito 11:35:00 10/27/03 (0)
    Hello Maciej
    thanks, i thought about this......look at this article:
    http://vt52.com/info/measure/6s45/6s45.html
    but do you think that the earlier cutoff at 10Khz would be reduced/eleminated?
    Here i post my schematic:
    the schematic looks like this:
    http://www.republika.pl/diytriode/EC8010_GM70.gif
    but i use 6c33 triode instead of GM70. The 6c33 runs with Ra=600Ohm experience electronics high quality transformer:
    http://www.experience-electronics.d...rager.htm#a6c33 (A6c33)
    Ub~200V, I~175mA
    At the grid of the 6c33 i get a cutoff beginns from 10Khz with 1-2dB! I need help.
    The 6c45 runs with 17ma, 175 at anode, interstage is LL1660/18mA in 1:1,25.
    
    Thanks Haukito

This post is made possible by the generous support of people like you and our sponsors:

Sonic Craft

You can not post to an archived thread.