I recently purchased a matched quad set of these tubes to replace a quad of Rubys where one of the tubes had failed. There are two tubes per channel.
The failed Ruby simply stopped conducting current, so I sprung for a set of Gold Lions.
When the amp is first turned on and the delay circuit finally kicks in, one channel of the amp produces a moderate hum for about 3 seconds, then it promptly goes away. The amp plays fine after that initial hum. Bias appears normal.
I swapped tubes between the left and right channels and the problem also follows the tubes, so I suspect that one tube has a problem.
Any thoughts?
Thanks in advance.
The failed Ruby simply stopped conducting current, so I sprung for a set of Gold Lions.
When the amp is first turned on and the delay circuit finally kicks in, one channel of the amp produces a moderate hum for about 3 seconds, then it promptly goes away. The amp plays fine after that initial hum. Bias appears normal.
I swapped tubes between the left and right channels and the problem also follows the tubes, so I suspect that one tube has a problem.
Any thoughts?
Thanks in advance.
Last edited:
Maybe one KT simply needs a bit longer to fully heat up and conduct properly. But 3s is too long for such a anomaly. Another option would be a heater cathode insulation failure. When not heated, the isolation failes and when heated up, the heater expands in the cathode tube a bit and it's gone. Last thing would be anomalies in the building up of the depletion region due to mechanical issues or gas inside the tube. It doesn't matter how you turn it....there is a failure in the tube.
I suggest to simply return the tubes where you bought them. I never bought any KT88 and when I would take the cheapest (because I can return them when they don't reach the specs) but when I hear the name "Gold Lion" I assume these weren't the cheapest KT's. So just return them
Regards
I suggest to simply return the tubes where you bought them. I never bought any KT88 and when I would take the cheapest (because I can return them when they don't reach the specs) but when I hear the name "Gold Lion" I assume these weren't the cheapest KT's. So just return them
Regards
I've heard rumors the New Sensor manufactured KT88's bias can drift wildly during warm-up. You said the amp is correctly biased and sounds fine. Is there a way to monitor the bias current as the tubes are warming up? It sounds like the mentioned tube is drawing too much current during start-up, causing the hum. Before returning the tubes, see if there's a way to prolong the delay circuit to 10 seconds or longer.
Thanks. The delay is currently set to 55 to 60 seconds. I would think 55 to 60 seconds would be more than long enough.
I only watched the bias with a digital meter and it did not vacillate wildly, but the affected side does tend to drift a little on one of the tubes, so that has me thinking that the tube is flakey.
I did wait until the system warmed up to adjust the bias and then rechecked every hour or so to fine tune it.
It seems like only one tube will vacillate after it warms up. The others show a very steady reading.
I'm measuring the bias current by looking at the potential difference across a 10 Ω cathode resistor. So, it is the actual difference across that resistor that fluctuates.
Another Gold Lion Bites the Dust.
Second tube has now gone bad. This one is microphonic when you tap lightly on it. It makes a cracking noise and it is definitely the tube.
I would hope this is simply a bad batch, but two tubes out of four is not a good indication for quality.
Has anyone else had or know of recent quality issues with Genalex KT88?
The Chinese Rubys at least lasted a year of more before one went belly up. The Gold Lions haven't made it two weeks before 50% of them failed.
Second tube has now gone bad. This one is microphonic when you tap lightly on it. It makes a cracking noise and it is definitely the tube.
I would hope this is simply a bad batch, but two tubes out of four is not a good indication for quality.
Has anyone else had or know of recent quality issues with Genalex KT88?
The Chinese Rubys at least lasted a year of more before one went belly up. The Gold Lions haven't made it two weeks before 50% of them failed.
Hi,
Last year, I bought 3 Gold Lion KT88 for my tubelab Simple SE.
On checking the Bias after a little while, on one of them, the bias kept slowly rising and rising...
I had bought the third one as a backup, so I decided to use it instead and at the same time see if it could be a problem developping on the amp itself.
Well, would you believe it, that third one got real noisy after a few days...
Popping sounds you know...
Then I was really begining to think that it was the amp, did some tests and it was the tube that was bad, wow!
I decided to use 6550's instead, the Tung Sol ones, well, got pretty surprised how good the amp sounded with these, I'd say as good than the kt88. To my ears anyway. A bit brighter, but in a good way
they got only about 100 hres on them so far, I dont listen only to that amp, got the bug and build a few so far
,
They're very stable so far. Also I use EDCOR OPT on the Simple SE.
I must say though that I use 4 Gold Lion KT66 on a pair of QUAD II , and these are Rock solid with more than 300 hours on them. The bias didn't move at all. Well, 1Ma on one of them. Impressive.
Last year, I bought 3 Gold Lion KT88 for my tubelab Simple SE.
On checking the Bias after a little while, on one of them, the bias kept slowly rising and rising...
I had bought the third one as a backup, so I decided to use it instead and at the same time see if it could be a problem developping on the amp itself.
Well, would you believe it, that third one got real noisy after a few days...
Popping sounds you know...
Then I was really begining to think that it was the amp, did some tests and it was the tube that was bad, wow!
I decided to use 6550's instead, the Tung Sol ones, well, got pretty surprised how good the amp sounded with these, I'd say as good than the kt88. To my ears anyway. A bit brighter, but in a good way
they got only about 100 hres on them so far, I dont listen only to that amp, got the bug and build a few so far
,They're very stable so far. Also I use EDCOR OPT on the Simple SE.
I must say though that I use 4 Gold Lion KT66 on a pair of QUAD II , and these are Rock solid with more than 300 hours on them. The bias didn't move at all. Well, 1Ma on one of them. Impressive.
Cathode resistor is 10.1 Ω each. Plate voltage is about 505 VDC.
Voltage across the resistor is set at .550 VDC.
If my math is correct:
(.550V/10.1Ω) * 505V = 27.5 Watts.
That should be well within the safety margin of the data sheet, which states 35 Watts.
Even if the plate voltage were 515 VDC it still is only 28 Watts.
I have had similar failures with the Gold Lion as well. Similar voltage and bias setting of 55mA. Two had the Screen short out. Two others failed like a flash bulb. I use a 1 watt 10 ohm for bias measurement that acts like a fuse. So the only damage was to replace the resistor.
Been using my EH kt88's since then with no problems for many months.
Been using my EH kt88's since then with no problems for many months.
A few thoughts about this...
Just some things to consider...
- I handle 1000s of KT-88 Genalex and others every year. The problem with any individual owner report is the sample size is far too small to draw any conclusions.
- As a rule ALL KT-88s have a failure rate that is higher than tubes that are easier to build such as the 6L6 family. To the poster that said the EH has been problem free - do a search and you'll find that other people will say the exact opposite - they'll say the EH were unreliable but the GLs were fine.
- The failure rate on the Genalex KT-88s is about 1% or a bit less in my testing; it's even smaller when you talk about returns from the users, most of the tubes that are problems are caught during my testing and matching. The same is true with the EH.
- The most common failure mode on the Genalex is interelectrode shorts. A few will not have stable bias.
- The pin to socket contact is often the cause of difficulty. Often you'll see it posted that "after I swapped tubes things got better". The act of removing and reinstalling a tube will most often improve the contact integrity; the wiping action as the pins are installed in the socket is sometimes enough to get adequate contact and the "tube problem" goes away. Properly clean (and retension if needed) your octal sockets periodically and the number of "tube problems" drops dramatically.
- If the only time the tube makes noise is when you tap it, then IMHO there is nothing wrong with the 2nd Genalex you wrote about. Almost any tube will make some noise when tapped.
- One last thing - the "factory matched" tubes are rarely if ever well matched. My experience is that 3 of a quad will be well matched but the 4th will often be way out. Also, "factory matched" tubes are matched for plate current only. Yet in most gear we measure/check bias by measuring cathode current (plate and screen current combined), so there is a problem from the start. Put mismatched tubes in many amps and you will have problems, especially if you have a tube that is passing more current that it's "matched" mates. One tube can easily be overworked while the others loaf along. That overworked tube is much more likely to fail. And of course we blame the tube - but in this case the tube is the victim, not the perpetrator!
Just some things to consider...
Jim, thanks for taking the time to answer.
As an electrical engineer I am aware of statistics, so I would not and have made a claim that what I saw was a trend. However, when two of the four tubes you buy have a failure mode and they are not the same type of failure, it raises some questions.
Similar failure modes would give credit to what is called "special cause" as a reason for a failure, probably due to variation in materials, processes, etc.
In this case my initial concern was if there was a general cause that may be more systemic.
There is no way to assign statistical analysis to determine that because we lack the type of data you have, but there may be an underlying issue that could explain it if the failure rate you see of 1% suddenly jumped to 5% or some other number.
I guess I am trying to stick my finger into the wind to see if there might such a change in production, but I also understand such assumptions must be drawn with grain of salt - as they say.
Incidentally, I have swapped tube positions and reseated tubes as you suggested and I have clearly narrowed it down to two specific tubes out of the set. One exhibits a hum at startup and also has drifting bias. The other produces popping and cracking sounds when lightly tapped, not when rocked in the socket.
I would like to replace this set with another Gold Lion set if there is no evidence of a systemic issue. Perhaps we should talk?
Last edited:
Keep in mind that these tubes are essentially hand made. So there is some process variation that occurs no matter what.
I try to weed out any problems by doing the matching personally and doing it at real world type voltages. For instance, the KT-88s run at 465 volts on the plate and screen with -50 on the grid. That's pretty close to where many amps run them. So I see more failures here on the bench than the end user sees by a good margin.
BTW, what are the production dates of the tubes you have? There was one production run maybe 3-4 years ago that seemed to have more than its share of issues, but I suspect those tubes are out of stock for most guys.
Feel free to drop me a note if I can be of help!
Closest thing I can find is a number, which I assume is some form of unique serial number:
7DB401A8290000A8
I don't know if that means anything.
Sadly no - that number is from New Sensor's "matching" setup.
The date will be on the tube itself, a four digit number; it's printed right on the tube glass in the same gold color as the rest of the lettering.
2010, 6th month production (June).
That's WAY after the one batch I was referring to as a bit sub-par.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.