The supply of NOS seems to be low/dwindled therefore not cheap. I have seen fake/scam websites appearing in google searches selling tubed so be careful who you buy from; some of the scammers hijack photos from google images/etc. There are a few big tube dealers in the USA that might be your best bet. Vacuumtubesinc, vacuumtube.net, vacuumtubes.com (SND). In the UK langrex.
If you're on a tight budget perhaps consider a redesign for a cheaper tube e.g. 6P41S.
If you're on a tight budget perhaps consider a redesign for a cheaper tube e.g. 6P41S.
Just a heads up new Tung Sol 7591 does not have the full ability to deliver anode current as "real" NOS 7591 from the past do.
The charts on the new 7591 confirm this reality.
Same only more for the JJ version.
So best to temper expectations.
The charts on the new 7591 confirm this reality.
Same only more for the JJ version.
So best to temper expectations.
I tried RCA, Sylvania and GE. Toss up with RCA and Sylvania. GE does not sound good. Most tube dealers have no stock. But N9TEW Tubes Sales has USA made 7591 for $60 each. They have JJ for $18 each.
What is the application for these tubes.
Is it a vintage amplifier and if so what model?
Is it a DIY build?
The JJ 7591 are significantly lower in anode current (max power output) than a NOS 7591 tube and will not be suitable for some applications.
See "Eico ST40" thread
Is it a vintage amplifier and if so what model?
Is it a DIY build?
The JJ 7591 are significantly lower in anode current (max power output) than a NOS 7591 tube and will not be suitable for some applications.
See "Eico ST40" thread
Even the EHs had trouble delivering anode current - unless the screen is cranked up so far as to over dissipate it. I’ve got a quad I’m not doing a whole hell of a lot with now.
Often, when amplifiers do not deliver the output power that you expect, there are a number of factors:
1. Output transformer insertion loss: Even a 0.5 dB insertion loss is noticeable.
If the Output Tubes actually deliver 10 Watts to the primary; the output transformer secondary only delivers 8.9 Watts (-0.5dB)
Tube datasheet output power specs do Not include the output transformer's insertion loss . . .
You control that loss when you select your output transformer.
2. Amplifier ratings: sometimes rated output power is at 1%, 2%, 3%, 4%, or 5%, THD
Even though the amplifier's rated distortion is 0.1%, 0.2%, 0.3%, 0.4%, or 0.5% THD
Big difference.
3. Your power mains voltage, and your power mains crest factor,
That causes the amplifier B+ to be less than what it was at the manufacturer's factory.
4. You are measuring at some frequency far lower than 1 kHz, or are measuring at some frequency far higher than 1 kHz.
5. You are using a cheap DMM to measure AC Volts, and its frequency response falls off lower than 1 kHz.
Cheap DMMs do not measure True RMS, so if there is any THD from the amplifier, the reading will not be true.
Trust but Verify
1. Output transformer insertion loss: Even a 0.5 dB insertion loss is noticeable.
If the Output Tubes actually deliver 10 Watts to the primary; the output transformer secondary only delivers 8.9 Watts (-0.5dB)
Tube datasheet output power specs do Not include the output transformer's insertion loss . . .
You control that loss when you select your output transformer.
2. Amplifier ratings: sometimes rated output power is at 1%, 2%, 3%, 4%, or 5%, THD
Even though the amplifier's rated distortion is 0.1%, 0.2%, 0.3%, 0.4%, or 0.5% THD
Big difference.
3. Your power mains voltage, and your power mains crest factor,
That causes the amplifier B+ to be less than what it was at the manufacturer's factory.
4. You are measuring at some frequency far lower than 1 kHz, or are measuring at some frequency far higher than 1 kHz.
5. You are using a cheap DMM to measure AC Volts, and its frequency response falls off lower than 1 kHz.
Cheap DMMs do not measure True RMS, so if there is any THD from the amplifier, the reading will not be true.
Trust but Verify
When a different set of tubes delivers the expected 25 watts with a given supply voltage and output transformer, and the 7591’s struggle to give me 15 it’s the tubes, not the transformer or the way voltage is being measured. I was asking for less than 200 mA of peak anode current and 7591’s should be able to do that without running the screen all the way up to 400.
Probably less than the specified cathode emission.
High screen voltage to get enough plate current, may strip away the cathode coating with time.
I remember there were some RF power tubes that can self sustain the cathode heating after the high plate current begins.
Do not attempt to do this with Audio Power tubes.
High screen voltage to get enough plate current, may strip away the cathode coating with time.
I remember there were some RF power tubes that can self sustain the cathode heating after the high plate current begins.
Do not attempt to do this with Audio Power tubes.
See "Eico ST40" thread posting on 2024-07-12 11:44 pm.
Current production 7591 tubes do not deliver the full anode current of NOS 7591s.
The result is except in lightly loaded circuits current production 7591 tubes deliver less power output sometimes substantially less than NOS 7591.
See chart below.
Really I feel current production 7591s should have a different part number as whatever they are they are not a 7591 by the 1960s data sheets.
Lots of people are tearing out hair wondering why their vintage amplifier has low power with a "new" set of 7591 tubes.
Myself was included in the hair pulling until I measured the anode current curves VS G2 and G1.
It then was very clear that full power was not going to happen with current production 7591 except in lightly loaded amplifier designs.
To compare a little 6BQ5 with 250V on its screen can deliver more anode current than a JJ7591 can with 300V on its screen.
No I am not saying to plug in 6BQ5s in place of 7591!
Attachments
Last edited:
That looks about right for the EH, too. No, maybe you can’t just drop in 6BQ5s into a 7591 circuit but you could build just as big an amplifier with them as you could new 7591’s.
A set of 4 6BQ5 (two in parallel on each side as P-P) delivers pretty much what a pair of old school 7591s delivered in terms of performance and with lower screen voltages.
How about some JJ KT77 tubes. Rated for 2x the current, similar gm, and more Watts and similar Vg2. Socket would need a couple of pins re-wired. Main issue could be the height. Could maybe lower the socket and/or get out the Greenlee punch for the "roof". Better curves too:
Last edited:
Don't forget that the 7591, like the 7868, were chosen for their compactness/power ratio and that it is generally impossible to physically substitute them for something else.
I use JJ and Tungsol in 4 or 5 of my amps, I have never been able to listen to a quartet of 7591 or 7868 NOS in good condition so I have no means of comparison but what comes out of my 7591 me fits perfectly after a few minor modifications which are widely documented on the internet.
I use JJ and Tungsol in 4 or 5 of my amps, I have never been able to listen to a quartet of 7591 or 7868 NOS in good condition so I have no means of comparison but what comes out of my 7591 me fits perfectly after a few minor modifications which are widely documented on the internet.
This I agree with as far as plug in replacements go including what is currently being sold and called 7591s.
The 5881 will physically fit in nearly all 7591 amplifiers however the bias and driving differences require considerable changes to the driver/bias circuits.
I have had good success replacing the 7591 with 5881 and after extensive driver/bias circuit changes got back to the same amplifier performance levels with NOS 7591.
Not so with the JJ7591. There is no way in my experience to overcome the lack of anode current available in a JJ7591 compared to a NOS 7591.
You end up with a little over 1/2 the normal power output in amplifiers that load the 7591 at the normal 6600 ohms P-P.
I would love to see a effective work around for the lack of anode current with JJ7591s. Can you please provide some links or searchable text for me to read up on these modifications?