Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum


Message Sort: Post Order or Asylum Reverse Threaded

CCS Issue

66.177.177.194

Posted on December 26, 2016 at 09:20:25
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
I modeled this CCS in LTSpice and voltage and current are right on the money when I fired this thing up. Seems to perform well and doing its job.
However, I stumbled on an issue when I fired up the amp with a 1volt signal at the input. The 150v at the plate dropped to 108v upon starting up the amp, when I removed the signal, the plate voltage stayed the same, it would never settle back to 150v. If I start the amp up cold with no signal again, everything is fine. I remember reading something about using a diode across the the CCS but don't remember where to put this.



 

Hide full thread outline!
    ...
RE: CCS Issue, posted on December 26, 2016 at 10:21:30
Chip647
Audiophile

Posts: 1675
Location: The South
Joined: December 24, 2012



12 volt zener from source to gate, pointed at gate.

Also, you probably damaged your MOSFET, just replace it.

 

RE: CCS Issue, posted on December 26, 2016 at 10:25:56
Tre'
Industry Professional

Posts: 12087
Location: So. Cal.
Joined: February 9, 2002
What device are you using for the CCS?



Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: IXCP10M90S, posted on December 26, 2016 at 12:17:29
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
.

 

RE: IXCP10M90S, posted on December 26, 2016 at 13:03:54
Tre'
Industry Professional

Posts: 12087
Location: So. Cal.
Joined: February 9, 2002
I have no experience with the 10m90s.

I use the 10m45s and sometimes the dn2540.

Have you replaced the 10m90s?

BTW Neither the 10m45s or the 10m90s is a MosFet.

Tre'


Have Fun and Enjoy the Music
"Still Working the Problem"

 

RE: IXCP10M90S, posted on December 26, 2016 at 13:53:33
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
I did replace the 10M90s. Somehow the input voltage is setting up the current and it wont revert back.
This is the same regulator as the 10M45s only a higher voltage rating.
It's dissipating about 6.5 watts and rated for 40
It was early....I know its not a MOSFET, I only remember reading about the diode protecting a CCS, must of been a MOSFET ckt and not the regulator.

 

RE: CCS Issue, posted on December 26, 2016 at 15:31:18
Paul Joppa
Industry Professional

Posts: 6687
Location: Seattle, WA
Joined: April 23, 2001
Many years ago (10? more?) there were survival issues with the 10M90S. I was only peripherally involved at the time, but it was never really resolved - I think trying one chip after another until a survivor was found, was the best solution. Not really satisfactory!

Speculation at the time was oscillation; the chip does have an enormous transconductance. Injecting a signal could trigger misbehavior, especially if something is pushed out of its linear range. At very high frequencies, minor changes in layout or lead dress can lead to unexpected oscillation.

I can only find the 6E5P as a Russian version of the 6E5 tuning indicator tube - is that the right tube? If so, 27mA into a 6E5 seems high since the rated target current is 4.5mA. That could lead to clipping which can also trigger oscillations.

 

6E5P, posted on December 26, 2016 at 15:52:25
Maxamillion
Audiophile

Posts: 820
Location: New Jersey
Joined: May 26, 2006
Here's the datasheet, Paul. Excellent driver tube.

 

RE: CCS Issue, posted on December 26, 2016 at 16:09:29
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
The 6E5P is a Russian tube capable of 8.3watts dissipation
Ik max is 100ma
Absolutely no oscillation I can see on a scope.
Maximillion (Magz) used this tube to drive a 833, it's a nice tube and behaves much better than pentodes I've used.

 

RE: 6E5P, posted on December 26, 2016 at 16:17:42
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
Maxamillion, what does the choke to ground do in lieu of a resistor in your amp?
I don't need grid current so I chose the CCS, I will say this has been one of my favorite tubes so far for my amp. It drives the 300B really nice with no coloration. Just got to figure out what's going on when I turn on the amp with a signal.
I might try your diode bias scheme next, might correct my little issue.

 

RE: 6E5P, posted on December 26, 2016 at 16:36:02
Maxamillion
Audiophile

Posts: 820
Location: New Jersey
Joined: May 26, 2006
That's a Lundahl grid choke. Just a substitute for the typical grid resistor. It gives a low impedance for DC (5K) together with a very high impedance for the music signal. It's not critical to the operation of the circuit, I just thought it would be a nice enhancement vs a resistor.

 

RE: Magnequest BCP16ni, posted on December 26, 2016 at 18:12:48
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
Never thought about putting one at the input tube.
I have a couple magnequest bcp16 nickel chokes lying around I might try.

 

RE: 6E5P, posted on December 27, 2016 at 13:26:12
Paul Joppa
Industry Professional

Posts: 6687
Location: Seattle, WA
Joined: April 23, 2001
That's different tube, all right!

I noticed the high transconductance (30,500 microMhos!), so oscillation is still a candidate. There is no grid stopper shown for the 6E5P, which makes oscillation very probable.

 

RE: CCS Issue, posted on December 27, 2016 at 13:35:20
Paul Joppa
Industry Professional

Posts: 6687
Location: Seattle, WA
Joined: April 23, 2001
Sometimes you need a really high frequency scope (400MHz?) and sometimes you must be very careful to not affect the tube by the measurement. It's very difficult to find intermittent, very high frequency oscillations. The only person I ever knew who had it down was the late John "Buddha" Camille.

I myself have neither the necessary measurement gear nor the expertise, but I have solved problems by installing an appropriate grid stopper, at which point I no longer have to care whether it was oscillation or not. :^)

 

RE: +1 on gridstopper, posted on December 27, 2016 at 18:17:03
deafbykhorns
Audiophile

Posts: 652
Location: Florida
Joined: October 17, 2003
.

 

RE: 6E5P, posted on December 28, 2016 at 19:00:28
twystd
Audiophile

Posts: 2559
Location: Austin,Texas
Joined: December 9, 2000



Paul I highly recommend for you to look at this tube. The 6e5p, and it's near twin (pinout different slight differences in Rp and Mu) the 6e6p-e, or better yet the 6e6p-dr. Here's the curves done by Jim Hagerman for the 6e6p-e, the divisions are 50v, and 5ma. I've read a post by Jim claiming these two tubes are the most linear he has ever tested. It's a tetrode and has g2 strapped with 100 ohm resistor. I've used it as a driver and have been highly impressed. I've also included some data sheets as well.

twystd

 

RE: 6E5P, posted on December 29, 2016 at 18:34:52
Paul Joppa
Industry Professional

Posts: 6687
Location: Seattle, WA
Joined: April 23, 2001
Thanks for the data - much better than what I found in a quick search.

 

RE: IXCP10M90S, posted on January 8, 2017 at 10:07:58
PakProtector
Audiophile

Posts: 11282
Joined: May 14, 2002
That is all in the heat sink size. anything over a Watt bare is pushing things.
cheers,
Douglas

Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.

 

RE: IXCP10M90S, posted on January 8, 2017 at 13:44:45
LinuxGuru
Audiophile

Posts: 537
Location: European Union
Joined: November 11, 2008
> BTW Neither the 10m45s or the 10m90s is a MosFet.

It is depleted mode mosfet (possibly BJT-cascoded) with built-in gate protection.

 

Page processed in 0.026 seconds.