SET Asylum

Single Ended Triodes (SETs), the ultimate tube lovers dream.

Return to SET Asylum


Message Sort: Post Order or Asylum Reverse Threaded

300B - Is this thermal runaway ?

49.37.68.181

Posted on October 26, 2021 at 06:46:35
Burney
Audiophile

Posts: 103
Joined: February 16, 2004
I have been trying to build the WE 91 clone based essentially on Joe Robert SP article.

I am using NOS 6SJ7 metal can and Shuguang Black Treasure 300B. The OPT is EDCOR CXSE25-8-3K.

All resistors are as per Joe's publication except the 300B Rk which is 1k-10w (Joe - 880R - 50w) and the grid leak resistor which is 220k (Joe - 390k).

The PSU in a 440VAC > 5UV > 10H > 36uF > 8H > 100uF > OPT. From the 8H Choke onwards it branches of to separate L & R rails.

The voltage readings are :
At rectifier 5U4 = 420 VDC
After first Choke = 416VDC
After Choke = 413 VDC
300B Plate = 394 VDC
300B Cathode = 55 VDC
Grid to Cathode = -50 VDC


6SJ7 Anode = 175VDC
6SJ7 Screen = 78 VDC
6SJ7 Cathode = 2.30 VDC

These measurement vary just a little based on wall voltages at the time of day.

The filaments are powered by a LM317 based DC regulator which takes about 6 seconds to reach the set 5 VDC.

The rig fires up fine and with a CDP plugged-in I can hear good music. No hum. No noise.

The problem arises when I crank up the Vol pot. At one point I just hear a "Dhap" and music goes off. 300B grid to ground goes up to almost 275 VDC and the Cathode voltage similarly climbs to about 120 VDC before I switch off.

I have rechecked all wiring. Interchanged the 300B and the driver 6SJ7. I don't have reason to doubt the EDCOR CXSE25-8-3K as DCR's for both measure equal. In fact they were being used in an EL34 SE type amp very successfully.

At this point I am at a loss to figure out what is happening here. Would be grateful for any pointers. I just have a good quality DMM and a Variac with me.

Need to add that both the 300B sounded fine and measured as per schematic specs in an ebay PCB that employed 6SN7 is a cascade type design. They have about 10 hours of use on them.

 

Hide full thread outline!
    ...
RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 07:34:16
Tre'
Industry Professional

Posts: 15718
Location: So. Cal.
Joined: February 9, 2002
replace the coupling cap.

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

 

RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 08:09:26
Burney
Audiophile

Posts: 103
Joined: February 16, 2004
Forgot to add, I did change the coupling caps. No relief.

 

RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 08:27:47
Tre'
Industry Professional

Posts: 15718
Location: So. Cal.
Joined: February 9, 2002
wait, you have two amps and they are both doing this?

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

 

RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 08:47:34
Burney
Audiophile

Posts: 103
Joined: February 16, 2004
I had this 300B ebay PCB to try out. Wired up everything and used a few days. It just wasn't as magical as the EL34 SE I had been using for several years. The EDCOR OPT's were repurposed from the EL34 SE amp and they worked fine with the ebay 300B PCB. No thermal runaway. I also recall 300B cathode had higher voltage, about 72VDC whilst using the ebay PCB. B+ was almost same

Got ambitious and put together the WE91 clone. And here we are in this situation.

 

RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 08:51:16
Tre'
Industry Professional

Posts: 15718
Location: So. Cal.
Joined: February 9, 2002
Did you build two WE91s and are they both having the same problem?

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

 

RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 09:11:15
Burney
Audiophile

Posts: 103
Joined: February 16, 2004
No. Just the current project in a makeshift chassis.

 

RE: 300B - Is this thermal runaway ?, posted on October 26, 2021 at 10:53:20
Tre'
Industry Professional

Posts: 15718
Location: So. Cal.
Joined: February 9, 2002
I would look for parasitic oscillation in the driver stage.

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

 

RE: 300B - Is this thermal runaway ?, posted on October 27, 2021 at 00:24:10
Burney
Audiophile

Posts: 103
Joined: February 16, 2004
Thanks. I installed range of stopper resistors from 220R upto 10k. Both at the driver and the 300B. No improvement

Secondly, as I understand the grid voltage at idle should be much higher than -50vdc that I am measuring right now.

 

RE: 300B - Is this thermal runaway ?, posted on October 28, 2021 at 10:39:23
Tre'
Industry Professional

Posts: 15718
Location: So. Cal.
Joined: February 9, 2002



Yes, if the tube is healthy with about 340 volts across the tube and only 50 volts bias there should be some 240ma of plate current flowing. That would melt the plate.

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

 

RE: 300B - Is this thermal runaway ?, posted on October 28, 2021 at 13:10:41
Paul Joppa
Industry Professional

Posts: 7152
Location: Seattle, WA
Joined: April 23, 2001
A few ideas re: "300B grid to ground goes up to almost 275 VDC and the Cathode voltage similarly climbs to about 120 VDC before I switch off."

If this is accurate, it seems likely the grid resistor is not connecting the grid to ground. Could be a cold solder joint or a bad socket connection/corroded grid pin on the 300B. Both might make contact when cold, but open up when warm - possibly triggered by movement when adjusting the volume control.

The other alternative is a bad coupling cap between the two stages. What cap are you using, in particular what is the voltage rating?

The low current is a separate issue. Possibly an exhausted tube, especially if it has been running excessive current reccently.

 

RE: 300B - Is this thermal runaway ?, posted on October 29, 2021 at 13:26:38
Burney
Audiophile

Posts: 103
Joined: February 16, 2004
Thanks Mr Joppa for the pointers.

I should have added in my OP that I have rechecked all solder joints, caps etc several times. Coupling caps tried included Mundorf Al oil 450v, Soviet K75 0.1uf 500v and some Arcotronics FF.

The 300B tubes work fine in an ebay PCB module that uses 6SN7's in cascade. In that board Vk is about 70vdc with an 1k Rk.

For 300B filaments I am using an ebay module that uses 1 qty of LM317. This ramps up the 5vdc slowly. Could this module be causing this grief? Current starvation? Filament voltage is stead at 5vdc as set.

Would it be helpful if I remove this regulator module and go AC or a simple bridge+caps arrangement for the filaments?

Thanks once again.

 

RE: 300B - Is this thermal runaway ?, posted on October 29, 2021 at 16:59:33
Paul Joppa
Industry Professional

Posts: 7152
Location: Seattle, WA
Joined: April 23, 2001
As you have tried 3 different caps, that seems unlikely to be the problem. But the grid is being raised 275v above ground, and there must be a reason. Several candidate reasons (oscillation, open grid-ground resistor or connections, leaky coupling cap) have been covered in this thread, and rejected for reasons - it may be necessary to question those reasons more deeply. Or come up with another cause, such as a short from B+ to grid

I wouldn't worry about the low current draw until that issue is resolved.

 

RE: 300B - Is this thermal runaway ?, posted on October 30, 2021 at 15:52:24
33dot33rpm
Audiophile

Posts: 56
Location: Nor Cal
Joined: June 8, 2006
Can you supply more info regarding the problem?

When the sound cuts off (SCO), is it for both channels?
Have you tried to run the amp since?
When SCO what is the voltage at the 100uF cap?
Are you using one LM317 regulated DC supply for both 300b?
It is strange that the 300b grid goes to 275VDC when SCO, are you sure that is the correct reading (grid to ground) and you are probing the correct pin (pin 3)?

Something else is not adding up with your build. If you are sticking strictly Joe's circuit (except for 300b rk), with B+ at 413Vdc and 300b rk = 1K typically the cathode voltage should be between 65 to 70 volts while you are seeing 55 at idle. Can you test the Edcor OPT primary impedance?

 

Page processed in 0.026 seconds.