richwalters said:
Hi there.........I still think it's a close run thing.....you might have tried the other rect tubes on a mains voltage high and you have problems......depends how much AC line variation you have where you are...
.Examine carefully
richj
My mains is nominal 120V but for the last 2 years has consistently been 124 whenever I've checked it. For this application my target B+ was 400V but in the end with my mains a tad high and with the Hammond PTX putting out slightly higher than spec'd voltage I got 430V which I think should still be OK, within 10%.
I did think about putting the old tube on a different circuit and varying voltage to it to see if there is a point where the problem starts or if it is just constant. Knowing myself though I probably won't get to doing that for quite a while. 😱
Just using mains might be interesting and would certainly be quick as I have a sort of PS test mule thingy I made with AC line in, fuse holder and terminal block to which I can connect whatever transformer I'm interested in at the time. Easy to use a different tranny or go straight from mains.
Hi there..........I threw the AC variations note in because I'm on the end of a power line i.e worst case .....however, with the number of tube vendors about all producing or copying each others tube designs......it's hard to assess todays bogey values as yesteryears NOS were far more conservative and reliable.
A variac is a good idea.......but all it needs is a darned good spike to flick a flash over. I've used an RC supressor across the tranny secondary; the cap voltage value needs to be generously rated. From my earlier experience, the US power line is pretty polluted.
See how you go......
richj
A variac is a good idea.......but all it needs is a darned good spike to flick a flash over. I've used an RC supressor across the tranny secondary; the cap voltage value needs to be generously rated. From my earlier experience, the US power line is pretty polluted.
See how you go......
richj
richwalters said:
Thanks for your input. I know my power is over voltage and I wouldn't be surprised if there was all kinds of "junk" on it as well. I've actually been considering a UPS with voltage regulator a'la Kofi Anan in another thread. Not due to any real issues, just as sort of a "control".
As for the flashing tube. I don't know how well I described what it was doing. It didn't make any noise and the flash didn't look like a spark. It appeared that the mica at the top would just flash white like a florescent light turning on.
The last time I used it it seemed that the flash "moved" from top to bottom on the tube and then just faded. During all this the tube seemed to be operating normally and it was definitely normal afterward as the voltage measured exactly the same as the first time I turned on the amp.
Bottom line at this point is that I won't worry about what was really happening with the old tubes and will keep my fingers crossed re: the Sovteks. If I have time I'll test the old one and if the Sovteks show issues I'll change the PS to have a smaller input cap and maybe go with NOS tubes.
SOVTEK
I am using a Sovtek 5u4g and it is doing the same thing. The powersupply cap is 20 uf then a 10 H choke and a 100 uf cap. The only load on the supply is a vltage divider comprised of a 220k in series with a 68k to ground. I am using a variac and the output voltage is 120 v . So I gues can pretty much rule out the cap. In my case is it because insuficient load ?
I am using a Sovtek 5u4g and it is doing the same thing. The powersupply cap is 20 uf then a 10 H choke and a 100 uf cap. The only load on the supply is a vltage divider comprised of a 220k in series with a 68k to ground. I am using a variac and the output voltage is 120 v . So I gues can pretty much rule out the cap. In my case is it because insuficient load ?
Hi tenderland,
Yes, the load is too light. This causes the second 100uF cap to be a problem. Disconnect it and the problem will go away.
There is a minimum load that will sustain the action of your choke, or it effectively disappears.
-Chris
Yes, the load is too light. This causes the second 100uF cap to be a problem. Disconnect it and the problem will go away.
There is a minimum load that will sustain the action of your choke, or it effectively disappears.
-Chris
Hi Jeff,
Uh huh. Disconnect the 100uF and try it again. Let us know what happens. Easy experiment.
-Chris
Uh huh. Disconnect the 100uF and try it again. Let us know what happens. Easy experiment.
-Chris
Sovtek
With the 100 uF cap disco it DOES NOT do it. I tried it 3 times to be certain. Thank you Chris
With the 100 uF cap disco it DOES NOT do it. I tried it 3 times to be certain. Thank you Chris
Hi Jeff,
The 100uF might be okay as long as the load current does not fall below the threshold current for that choke. A 40uF cap would be more in keeping with what I normally see in a well designed supply.
It's strange that these things were already known and spelled out and some current designers make the same errors. Counterpoint being a prime example of this. The problem is so many poor examples for people starting out to find and follow.
-Chris
The 100uF might be okay as long as the load current does not fall below the threshold current for that choke. A 40uF cap would be more in keeping with what I normally see in a well designed supply.
It's strange that these things were already known and spelled out and some current designers make the same errors. Counterpoint being a prime example of this. The problem is so many poor examples for people starting out to find and follow.
-Chris
UPDATE
I decided to rearange my configuration to make new hum measurements. I left the 100uF cap off AND it is still doing it, maybe before i could not see well due to ambient light but room is darker now I have another 5u4g tube(SOVTEK) but was saving that one fro the finished project. At this point I would say it is the tube. The only thing that is unknown is how it will react to the proper load as it is now
288K.
I decided to rearange my configuration to make new hum measurements. I left the 100uF cap off AND it is still doing it, maybe before i could not see well due to ambient light but room is darker now I have another 5u4g tube(SOVTEK) but was saving that one fro the finished project. At this point I would say it is the tube. The only thing that is unknown is how it will react to the proper load as it is now
288K.
Hi Jeff,
You may be chasing a non issue. I assumed your flash was a greater magnitude. Generally, if you can't see it in normal light or if you really have to look it's not bad. Still it would be interesting to see what a Tungsol, GE, RCA tube does.
What I've been seeing is circuits where there is 220uF hanging off a 6CA4 or the like.
You may be chasing a non issue. I assumed your flash was a greater magnitude. Generally, if you can't see it in normal light or if you really have to look it's not bad. Still it would be interesting to see what a Tungsol, GE, RCA tube does.
What I've been seeing is circuits where there is 220uF hanging off a 6CA4 or the like.
I'am having similiar issues with my 5R4wgb rectifiers. I have found that the flash over goes away if you leave the center taps of the transformer unconnected to earth. Doesn't seem to draw much current. I will be installing a simple time delay on the center tap earth lead to implement this.
I have no idea why this works, but it seems to. I know there is less current been drawn because my choke stops buzzing.
Shoog
I have no idea why this works, but it seems to. I know there is less current been drawn because my choke stops buzzing.
Shoog
Hi Shoog,
If you leave the centre tap unconnected you have created a standby switch. The supply will not deliver any real current until it is reconnected. A timed connection is a good idea since the rectifier's heater can warm up. Helps the rest of the tubes as well not to apply B+ until their heaters have warmed up. 30s should do it, 45s if you want to be ultra careful.
-Chris
If you leave the centre tap unconnected you have created a standby switch. The supply will not deliver any real current until it is reconnected. A timed connection is a good idea since the rectifier's heater can warm up. Helps the rest of the tubes as well not to apply B+ until their heaters have warmed up. 30s should do it, 45s if you want to be ultra careful.
-Chris
Sounds great, thanks for the explanation. I have some nice high current relays which should do service.
shoog
shoog
Sherman......hi there....I don't think youv'e mentioned the mains tranny type/size/ VA you are using for your KT mono blocks....with the 5U4 rects..
One of my 807 amps still has orig 5U4 Kenrad for 25 yrs without failure....and is in operation most weekends with wind instrum..
I know that the mains tranny is somewhat <soft> i.e with high resistance secondaries but rated at 450-0-450. I'm feeding into 4uF paper and 7.5H choke....with 2x 50uF electy on B+ o/p.
richj
One of my 807 amps still has orig 5U4 Kenrad for 25 yrs without failure....and is in operation most weekends with wind instrum..
I know that the mains tranny is somewhat <soft> i.e with high resistance secondaries but rated at 450-0-450. I'm feeding into 4uF paper and 7.5H choke....with 2x 50uF electy on B+ o/p.
richj
whoa guys! let me get this straight... if the datasheet says 40uF, I cannot put 100uF cap after the rectifier?
But I can use >40uF in the next section? (CRCRC)
I am using NOS chelmer 6x4... and I want to make sure it doesnt die before I do.....
But I can use >40uF in the next section? (CRCRC)
I am using NOS chelmer 6x4... and I want to make sure it doesnt die before I do.....
richwalters said:
My PTX is a Hammond 273BX, 350-0-350, 150mA (IIRC) max DC. They get pretty warm but not egg-frying hot.
Hi there,
I implemented the time delay circuit today. Before the delay switches the earth in, the DC output is very low 8V on the 300V supply, and 47V on the 900V supply.
I still get a bit of a flash over on the 300V supply because the heater supply is a little low at 4.45V. Unfortunately this will have to stay as there just isn't space for another transformer. I would say that the situation is a lot better though.
Shoog
I implemented the time delay circuit today. Before the delay switches the earth in, the DC output is very low 8V on the 300V supply, and 47V on the 900V supply.
I still get a bit of a flash over on the 300V supply because the heater supply is a little low at 4.45V. Unfortunately this will have to stay as there just isn't space for another transformer. I would say that the situation is a lot better though.
Shoog
Hi there.......have you thought about a resistor bleeder across the relay to get the B+ up a bit before she switches in ? i.e a value so that 1/3 volts is already up ? I do have have some sympathy for the rectifiers having to switch a lump sum current....it's a dogs life. Perhaps I'm becoming a tube softie.
It's an old design habit....I do like conservative running my tubes ....unless a design calls for grand slam output.
richj
It's an old design habit....I do like conservative running my tubes ....unless a design calls for grand slam output.
richj
- Status
- Not open for further replies.