Hi all,
First-time poster here hoping for some guidance. I've done a search on my issue and thought I had it figured out, but alas...
So what I'm dealing with is JE Labs 300B monoblocks (schematic attached). About 4 months ago, I noticed that half of the filament on the right channel power tub was out. They were Electro-Harmonix 300Bs, about 10 years old, probably in the 12,000-15,000 hour range, so I figured the tube had just died. Replaced the tubes.
Then within a couple months, the same thing happened on a brand new tube. The amp still worked, but half the filament wouldn't light. OK, so it's probably the cathode bypass cap that's gone bad, right? Then half of the left channel 300B filament stopped lighting, and then finally the rest of the right channel went. Not filament at all on the R 300B.
Now here's where things get odd. The R-channel 100uF cathode bypass cap is actually the one that I replaced about 6 years ago, and it tests out just fine. The L-channel bypass cap, though, is all over the place, clearly on its last legs. The 880Ohm cathode resistors also test accurately on both sides.
So I'm perplexed at the moment. Could it be the hum balance pot? And how exactly would one go about testing that to be sure? Or something else? I'm not a very experienced trouble-shooter, so please forgive the uncertainty.
First-time poster here hoping for some guidance. I've done a search on my issue and thought I had it figured out, but alas...
So what I'm dealing with is JE Labs 300B monoblocks (schematic attached). About 4 months ago, I noticed that half of the filament on the right channel power tub was out. They were Electro-Harmonix 300Bs, about 10 years old, probably in the 12,000-15,000 hour range, so I figured the tube had just died. Replaced the tubes.
Then within a couple months, the same thing happened on a brand new tube. The amp still worked, but half the filament wouldn't light. OK, so it's probably the cathode bypass cap that's gone bad, right? Then half of the left channel 300B filament stopped lighting, and then finally the rest of the right channel went. Not filament at all on the R 300B.
Now here's where things get odd. The R-channel 100uF cathode bypass cap is actually the one that I replaced about 6 years ago, and it tests out just fine. The L-channel bypass cap, though, is all over the place, clearly on its last legs. The 880Ohm cathode resistors also test accurately on both sides.
So I'm perplexed at the moment. Could it be the hum balance pot? And how exactly would one go about testing that to be sure? Or something else? I'm not a very experienced trouble-shooter, so please forgive the uncertainty.
12K+ hours is pretty good!
But sadly, the Electroharmonix (and Sovtek) 300Bs often fail like this. I have seen it in 1998 production, and later years.
It is caused by mechanical breakage of the filament. The Svetlana SV300B was bad also, for the same fault.
Using ac-heating does not help - the 300B's cold filament resistance is about 1/4 of the normal running value; the startup heating current can be correspondingly higher. When an all-in-one power trafo combined with vacuum rectifier is used, the available power at the filament winding is greater still, because the HV winding is effectively unloaded (rectifier still cold). So the filament gets a big jolt at startup.
EML versions seem to be better made, at high cost. They have a slow-warming filament to combat this failure mode. JJs 300Bs were very good in the late 1990s - maybe they still are.
But sadly, the Electroharmonix (and Sovtek) 300Bs often fail like this. I have seen it in 1998 production, and later years.
It is caused by mechanical breakage of the filament. The Svetlana SV300B was bad also, for the same fault.
Using ac-heating does not help - the 300B's cold filament resistance is about 1/4 of the normal running value; the startup heating current can be correspondingly higher. When an all-in-one power trafo combined with vacuum rectifier is used, the available power at the filament winding is greater still, because the HV winding is effectively unloaded (rectifier still cold). So the filament gets a big jolt at startup.
EML versions seem to be better made, at high cost. They have a slow-warming filament to combat this failure mode. JJs 300Bs were very good in the late 1990s - maybe they still are.
But the second set of tubes that exhibited the same issue were Gold Lion. Same amp, same side of the filament went out as in the EH tubes. I get that there are known issues with tubes, but it just doesn't seem likely that two tubes from different manufacturers will have filament failures in exactly the same spot. Perhaps, but it seems prudent to try to diagnose any circuit components failing before sticking another set of expensive tubes in and burning those out too.
The Gold Lion and EH 300B are both made by New Sensor in the Expo Pul plant in Saratov Russia.
I had a lot of Svetlana Spb 300B fail back in the early 2000s. (At least 8)
I have never had a JJ 300B fail, including ones of recent production.
Has your line voltage increased recently, changed the power cord or other arrangements that might make more inrush current available?
I would replace both cathode bypass caps with new ones, worth making sure they are the same, resist the urge to oversize.
I now use CCS based heating of various sorts including Rod's and have not had a problem with filaments failing.
I had a lot of Svetlana Spb 300B fail back in the early 2000s. (At least 8)
I have never had a JJ 300B fail, including ones of recent production.
Has your line voltage increased recently, changed the power cord or other arrangements that might make more inrush current available?
I would replace both cathode bypass caps with new ones, worth making sure they are the same, resist the urge to oversize.
I now use CCS based heating of various sorts including Rod's and have not had a problem with filaments failing.
Ah ha. To bad I didn't realize this until after the 90-day warranty. Maybe JJs are the way to go. I also had several sets of Svetlanas fail in the early '00s.
Nothing has changed in the power setup or anything with the building electricity, so I'm not sure I can spot a culprit there.
Angela audio supply, where I originally got all the parts for this amp in 1998, has replacement 100uF capacitors, but only at 50VDC and 450VDC; there's no 100VDC option as indicated in the schematic. Anyone have a recommendation for a good supplier of audio-grade capacitors?
Nothing has changed in the power setup or anything with the building electricity, so I'm not sure I can spot a culprit there.
Angela audio supply, where I originally got all the parts for this amp in 1998, has replacement 100uF capacitors, but only at 50VDC and 450VDC; there's no 100VDC option as indicated in the schematic. Anyone have a recommendation for a good supplier of audio-grade capacitors?
I think I lost 10 Svetlana's. All the same failure, the filament would actually move as it warmed. B+ was up due to the 5R4 fast filaments.
Arc and spark, eventually half a filament.
Sad, they sounded very nice!
Enter EH......2 failed in the first 3 months. The seller begrudgingly replaced them.
2 years later, one still alive. 2 literally went up in flames inside. Both lost vacuum.
Now I have Gold Lions. Hard to hear they come from the same factory as EH. But after seeing them, I already knew.........
Been 4 months......so far so good. I'll be pissed if they fail.
Should note that I added soft start to the B+ before using these.
BTW......I have one original engraved base WE 300B.
Only half the filament works........
I'm saving the 3 cases of NOS engraved base WE 300B's for retirement........kidding!
Ron
Arc and spark, eventually half a filament.
Sad, they sounded very nice!
Enter EH......2 failed in the first 3 months. The seller begrudgingly replaced them.
2 years later, one still alive. 2 literally went up in flames inside. Both lost vacuum.
Now I have Gold Lions. Hard to hear they come from the same factory as EH. But after seeing them, I already knew.........
Been 4 months......so far so good. I'll be pissed if they fail.
Should note that I added soft start to the B+ before using these.
BTW......I have one original engraved base WE 300B.
Only half the filament works........
I'm saving the 3 cases of NOS engraved base WE 300B's for retirement........kidding!
Ron
Ditto. I have several Sovtek 300B's where only half of the filament lights. New Sensor was aware of the problem at least 10 years ago. Their answer was to "buy our EH 300B, it has a better filament." NOT!!!!
I haven't bought any of their DHT's since, because they still fail in the same manner. It does seem to be batch related, since you can go for years without finding one, then see several in a short period of time.
I haven't bought any of their DHT's since, because they still fail in the same manner. It does seem to be batch related, since you can go for years without finding one, then see several in a short period of time.
Thanks everyone. I figured I'd test the bypass caps on the 6SN7 stage too, where I currently have Black Gates. It looks like one of those isn't giving a proper reading anymore, so I'm looking at replacing them I guess. Obviously Black Gates aren't available, but I seem to be having difficulty finding anything with the same values: 47uF/160V. Anyone have a suggestion for a good substitute?
I know they are not boutique audio caps but I almost exclusively use Evox Rifa/Kemet PEG124 caps for cathode bypass and most other uses. Check out Mouser they stock the full range. They are ideal for valve amps being axial and also have a 125°C maximum temperature rating. Also the other specs such as ESR etc. speak for themselves.
Cheers
Matt
Cheers
Matt
I have yet to have any of the EH Gold 300B tubes fail for any reason - and I see a reasonable number of them since I'm a tube seller.
I have no recent experience with New Sensor's DHT's at all since the rather blunt replies to my emails and phone calls about 10 years ago. I had been a fan of the Sovtek 6550 type tubes in my guitar amp builds, but the attitude I got when they wouldn't even replace a 3 week old tube because it had a partially open filament (the warrantee doesn't cover "blown" filaments) followed by a blunt refusal to replace what turned out to be bunch of filament failures in multiple amps over about a year convinced me to start using Chinese tubes, at least I could get the bad ones replaced.
I cracked open a Sovtek 300B to perform an autopsy years ago. The filament wire was not broken. It was simply not making contact with the support post where it had been crimped. It looked like the wire was coated with emissive material inside the crimp.
The early Sovtek 300B's also had a runaway problem at high plate voltages due to the fact that the filament was not covered by the grid at the top of the tube. This allows a direct electron path from the filament to the plate that tons of negative grid voltage can't control.
After years of not using any New Sensor products, about 6 years ago I was offered a box full of "pre auditioned" New Sensor octal output tubes (mostly 6550's) for $10 each. All had Tube Store matching stickers. I don't know if they were customer returns or not, but the EL34's all had "teeth marks" from the retainers used in some guitar amps. These tubes were all excellent and all are still operating. I have been using EH 6550's and KT88's in new builds.
It is possible that they have solved the problem in their newer DHT's, but I haven't tried any in recent years. The few DH amps that I have built have used NOS 307A's either in pentode configuration, or triode wired in place of a 300B.
I cracked open a Sovtek 300B to perform an autopsy years ago. The filament wire was not broken. It was simply not making contact with the support post where it had been crimped. It looked like the wire was coated with emissive material inside the crimp.
The early Sovtek 300B's also had a runaway problem at high plate voltages due to the fact that the filament was not covered by the grid at the top of the tube. This allows a direct electron path from the filament to the plate that tons of negative grid voltage can't control.
After years of not using any New Sensor products, about 6 years ago I was offered a box full of "pre auditioned" New Sensor octal output tubes (mostly 6550's) for $10 each. All had Tube Store matching stickers. I don't know if they were customer returns or not, but the EL34's all had "teeth marks" from the retainers used in some guitar amps. These tubes were all excellent and all are still operating. I have been using EH 6550's and KT88's in new builds.
It is possible that they have solved the problem in their newer DHT's, but I haven't tried any in recent years. The few DH amps that I have built have used NOS 307A's either in pentode configuration, or triode wired in place of a 300B.
As I've remarked earlier I have had no problems at all with JJ 300B over a span of 15 yrs or more and recommend them to all of my friends and former clients. Have not heard of problems even with recently purchased ones, I always recommend purchasing ones that have been burned in and matched and that seems to prevent problems. Quality has been very consistent over the years.
I prefer them sonically to anything except vintage WE and current EML.
I prefer them sonically to anything except vintage WE and current EML.
Eurotubes is here in Portland and the owners son is a member of our audio group.
He is going to bring over a matched Quad of the 300B for a listen. Should be interesting.
They have direct ties to JJ. So he knows about all the other factories.
I use their ECC99 as drivers.
I'm prob one of the few with PP amps.
Kevinkr built a nice P-P 300B amp, and published the build in a magazine. (VTV?) I don't know if he still has it.
I built one about 12 years ago. The 300B's were heated with individual 5 VAC CT transformers per channel with IRL's in series with the primaries. Within the first year 3 of the 4 Sovtek 300B's lost half of the filament, although one would work whenever it felt like it. That's the one I cracked open. After the 300B's died I put the amp in the closet.
A year or two later Stan (ESRC) offered me 4 used Sovtek 300B's for $100 (yeah $25 each) with a 10 day guarantee! The were obviously older than what I took out of the amp, but all 4 worked....and still do. The amp however does not. I built it as an experiment, using all junk box parts, but it turned out excellent, worked flawlessly for 5 or 6 years, and it's one of only two amps I saved from the early Tubelab days. One of the photoflash electrolytics in the power supply shorted. I stuck in a bigger fuse and powered up...toasting the diodes and IRL in the power supply. I plan on rebuilding it some day and using the same quad of really old Sovtek 300B's.
Yes, it was VTV! 
Long since moved along to a fellow who wanted them quite badly.
I use EML tubes in my current line stage design and plan more designs using them in the future.
It is always a question of budget, but I really like the EML tubes, they seem to be well made and quite linear IMVLE. So far most of my experience has been with the 20AM..
--------
Note that the JJ 2A3-40 is their 300B with 2.5V filaments and particularly for AC heated applications may be even better than their 300B.
Long since moved along to a fellow who wanted them quite badly.
I use EML tubes in my current line stage design and plan more designs using them in the future.
It is always a question of budget, but I really like the EML tubes, they seem to be well made and quite linear IMVLE. So far most of my experience has been with the 20AM..
--------
Note that the JJ 2A3-40 is their 300B with 2.5V filaments and particularly for AC heated applications may be even better than their 300B.
Off topic
I actually live in Vacuum Tube Valley, i.e. Lakeport California, where Charlie Kittleson operated out of. I met him a couple times a few years after I moved here in order to buy some tubes. He was an interesting fellow and we talked tubes, and also a bit more. As discussions progressed over time I realized he seemed to be haunted by some issues going on in his life and wore those feelings on his sleeve. I still didn't really understand the extent of his fragility until I noticed his shop was closed and didn't understand why. Out of respect for his memory I won't elaborate beyond that except to say that may he rest in peace.
I actually live in Vacuum Tube Valley, i.e. Lakeport California, where Charlie Kittleson operated out of. I met him a couple times a few years after I moved here in order to buy some tubes. He was an interesting fellow and we talked tubes, and also a bit more. As discussions progressed over time I realized he seemed to be haunted by some issues going on in his life and wore those feelings on his sleeve. I still didn't really understand the extent of his fragility until I noticed his shop was closed and didn't understand why. Out of respect for his memory I won't elaborate beyond that except to say that may he rest in peace.
I remember that I used your schematic from the magazine, and another schematic from this forum as starting points to make mine. Then I morphed those ideas together with my existing parts connection to make an amp without spending much money. It was one of the most dynamic amps that I have ever made, and I could not bring myself to cannibalize it once it died. I "borrowed" the power supply choke to build my 845SE amp, which is also sitting next to the "300Beast" on a shelf awaiting rebuild. The 300Beast remains intact except for the missing choke, and the 845SE is complete, and was working when stored, but has not seen power in at least 7 years.
I would like to rebuild the 300Beast completely, with some design changes. I can't find either of the original schematics since I have gone through several computers since then, and my magazine collection got pared down a lot when I moved.
I designed a universal driver board several years ago which would suit 300B's just fine, and the old toroid, silicon and photoflash caps that I called the power supply will be voted off the island. The new driver is an LTP feeding a second LTP with mosfet drivers to allow A2 and drive big tubes if needed.
funny, i hear this often. but how come no bias adj needed, when i swap with sovteks
true 300b would be off ,imo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.