Just a note-
The stock Knight KA95 uses many ceramic caps as coupling caps in the preamp section and tone controls. In stock configuration this amp could never sound great due to the ceramic caps. You would be better off with a new Yaqin MC-13S and a tube phono amp. The Yaqin uses quality polypropylene coupling capacitors. Not quite as good sounding as paper in oil, but still very good capacitors and a cut above Sprague Orange Drops.
Once the ceramic and other caps are replaced with Russian K40Y-9 paper in oil, this amp does have the capacity to sound world class. I like tone controls and built-in phono amp.
The stock Knight KA95 uses many ceramic caps as coupling caps in the preamp section and tone controls. In stock configuration this amp could never sound great due to the ceramic caps. You would be better off with a new Yaqin MC-13S and a tube phono amp. The Yaqin uses quality polypropylene coupling capacitors. Not quite as good sounding as paper in oil, but still very good capacitors and a cut above Sprague Orange Drops.
Once the ceramic and other caps are replaced with Russian K40Y-9 paper in oil, this amp does have the capacity to sound world class. I like tone controls and built-in phono amp.
JJ capacitor 500V- 550 volts for 9 Seconds
Can a JJ capacitor rated at 500 volts take 550 volts for 9 seconds? After the tube warm up period, the capacitor operates at 400 volts. I tried it once and all was ok. (I have not finished rebuilding the amplifier yet). I read Cornell Dubilier tests their 500 volt rated caps at 600 volts for 30 seconds.
Can a JJ capacitor rated at 500 volts take 550 volts for 9 seconds? After the tube warm up period, the capacitor operates at 400 volts. I tried it once and all was ok. (I have not finished rebuilding the amplifier yet). I read Cornell Dubilier tests their 500 volt rated caps at 600 volts for 30 seconds.
Last edited:
No bleeder resistors in the stock amp. A bleeder resistor would do little to lower the no load voltage during warm up & at 50 volts lower it would be a heater. Plus likely over load the power transformer B+ winding during operation. So, no option here.
I could install a time delay tube in the HV ground and automatically delay the B+, but really do not like that arrangement due to the brief momentary contact during closing (slight arcing). So, I need to install a latching relay as well. A lot of complication.
See an innovative device from rubli :
http://n.manet.nu/tubedoc/rubli-article.pdf
Available at :
A Start-Up Delay For Vintage Amplifiers
Check out page 8 of the Cornell Dubilier capacitor application guide http://www.cde.com/resources/catalogs/AEappGUIDE.pdf and part 2-3 on Nichicon Aluminium Capacitors guidelines. http://www.nichicon.co.jp/english/products/pdf/aluminum.pdf. In short, on large capacitors, surge overvoltages of no more than 50V will shorten the capacitor life due to leakage current. The formula is on the Rubycon application note, section 5-3 Product Information: Aluminum Electrolytic Capacitor Technical Notes/Capacitor, Power Supply Units RUBYCON CORPORATION
On smaller capacitors and/or higher overvoltages will cause the opening of the pressure vent (the capacitor will "explode"). 500V JJ capacitors are often overstressed in this way on guitar amplifiers and I never saw one explode, but they don't last very long.
On smaller capacitors and/or higher overvoltages will cause the opening of the pressure vent (the capacitor will "explode"). 500V JJ capacitors are often overstressed in this way on guitar amplifiers and I never saw one explode, but they don't last very long.
Knight KA95 too high voltage problem & solution
The not too popular Knight KA95 EL34 ultra linear amplifier is a duplicate of the Mullard 20 circuit and has a pair of excellent Chicago audio transformers. It is a worthy vintage amplifier with a non pedigree name.
It uses two Mullard GZ34 rectifiers that with todays higher line voltage supplies a scorching 495 volts to the EL34 plates & screens. The EL34 tubes show an orange color with the screens also seeing the excessive high voltage. New paper in oil coupling caps are not leaking lowering the tube bias.
I estimate about 28-30 watts plate dissipation on the EL34 tubes. The bias is set with series filaments on the preamp tubes at -37 volts- no higher bias voltage option here. Additionally, I did see a like new tested Mullard GZ34 arc during warm up. This is not good.
I installed a pair of 5R4GYB rectifiers. The 200ma increased current draw does not over load the transformer 5 volt winding. Now, the B+ voltage is 450 volts and the EL34s do not show any color on the plates.
Only problem is the 5R4GYB tube height does not allow the case cover to be installed. A solution is to buy a Hammond aluminum 546-1411CU 'clamshell' case. I used 1/2 of this case with the flanges and punched in two tube socket holes and installed the case on the Knight chassis. The rectifier tubes are now in a horizontal position and fit with the case installed. I did orientate the tube sockets so the rectifier filaments do not sag towards the plates. And, at 3100 volts PIV per rectifier tube, no chance of a tube arc over.
For what it is worth perhaps the Knight KA95 is not worth the changes with so many other vintage amps available, but at least these modifications go on record as a solution to the over voltage issue.
The not too popular Knight KA95 EL34 ultra linear amplifier is a duplicate of the Mullard 20 circuit and has a pair of excellent Chicago audio transformers. It is a worthy vintage amplifier with a non pedigree name.
It uses two Mullard GZ34 rectifiers that with todays higher line voltage supplies a scorching 495 volts to the EL34 plates & screens. The EL34 tubes show an orange color with the screens also seeing the excessive high voltage. New paper in oil coupling caps are not leaking lowering the tube bias.
I estimate about 28-30 watts plate dissipation on the EL34 tubes. The bias is set with series filaments on the preamp tubes at -37 volts- no higher bias voltage option here. Additionally, I did see a like new tested Mullard GZ34 arc during warm up. This is not good.
I installed a pair of 5R4GYB rectifiers. The 200ma increased current draw does not over load the transformer 5 volt winding. Now, the B+ voltage is 450 volts and the EL34s do not show any color on the plates.
Only problem is the 5R4GYB tube height does not allow the case cover to be installed. A solution is to buy a Hammond aluminum 546-1411CU 'clamshell' case. I used 1/2 of this case with the flanges and punched in two tube socket holes and installed the case on the Knight chassis. The rectifier tubes are now in a horizontal position and fit with the case installed. I did orientate the tube sockets so the rectifier filaments do not sag towards the plates. And, at 3100 volts PIV per rectifier tube, no chance of a tube arc over.
For what it is worth perhaps the Knight KA95 is not worth the changes with so many other vintage amps available, but at least these modifications go on record as a solution to the over voltage issue.
@sony6060 - it’s against the Forum rules to start multiple threads for the same project, as a result, your threads have been merged.Per the Rubycon formula, perhaps my capacitor life would be more normal of the capacitor being operated at the voltage rating limit due to once the amp is drawing tube current I see only 400 volts on the JJ capacitor that increases capacitor life. I think I will take the chance of the temporary 50 volts over spec. If the capacitor shorts, all I lose is a 6K 3 watt resistor vs possible rectifier tubes failure and fuse failure during a short at full B+ current. Thanks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.