The 6AQ5 is specified for audio amplifier service where is is rated for 12 watts of dissipation. It is also specified for TV vertical sweep (frame output) where is is rated at 10 watts. When these tubes were designed and rated many houses had only one TV set, and it saw a lot of use, hence the lower ratings for TV service.
The GE sheet I have shows 40 mA average cathode current, and 115 mA peak current. The Tung Sol sheet has the same numbers. The peak current rating is usually only shown on tubes for TV service, and is the important spec. It is the measure of cathode current that can be drawn without depleting the "electron cloud" surrounding the cathode.
The French datasheet you linked is for the short plate 6005 / 6AQ5W tube which was a military tube similar to the standard 6AQ5. About 1/3 of my "6AQ5" collection is this type of tube although a quick look doesn't reveal any French tubes, but many have lost their paint due to poor storage. The French data sheet is the first I have ever seen with positive grid curves. It is also a typical military type datasheet with test limits on it. The anode current listed on the bottom of the second page is one of those test limits. A good tube must measure between 33 and 57 mA when G2 and the plate are at 250 volts, and G1 is at -12.5 volts. They probably used the US type number on these tubes because these are not true 6AQ5 / EL90. The dissipation spec is only 11 watts. These may have been made for the military equipment market.
All of my tubes came from a military scrapper who pulled the tubes and other saleable items, stored them and scrapped the metal. There were over 100,000 tubes in the abandoned building when the bulldozers came. They had all been thrown randomly in boxes and were headed to a landfill when someone found mercury and halted the demolition of the building. Thst's when a friend bought the lot cheap for some saleable parts. I got all the tubes free for a few days work moving it all so the building could be torn down. It took me 5 years to sort through boxes like these. When it was all done, the 6005 / 6AQ5W was one of the more common tubes. At the time I put them all in with the other 6AQ5 / 6005 / 6HG5 / 6095 tubes, since there are some tubes with all 4 type numbers on them. I now know that there at least two different kinds, both of which will likely work just fine in any "normal" 6AQ5 circuit.
The $4 transformer says "10 W." There several taps for 25 and 70 volt lines, but the useful ones are common (one end), 70V / .0625W (CT) and 70V / 2.5 W (other end). I got a brand new one last week since I the $5 (before the sale) got me free shipping and was cheaper than shipping. It looks a bit different than the ones from the HBAC days. I hooked up to my audio oscillator and tested its ratio. The 4 ohm tap is 35:1 and the 8 ohm tap is 35:1, or roughly 5K ohms.
Lack of inductance is what blows output tubes, especially when those tubes are pushed beyond their ratings. Low primary inductance causes the transformer to saturate at low frequencies. This causes its already low inductance to drop toward zero. The output tube is now trying to drive a near zero load, raising the tube current, which pushes the transformer further into saturation. Red plate happens.
I think 5000 ohms is probably a bit low for already squeezed 6AQ5's, but i'll try it anyway, and try an 8 ohm load on the 4 ohm tap. I can also try bigger tubes.
Also: class A audio amp work is steady and predictable. The tube draws a large power and diverts a small amount to a load. Dissipation is easily measured. Switch-mode sweep work involves large input power and hard switching so little of that power stays in the tube. Before fancy 'scopes it was difficult to even measure dissipation. If anything goes a little wrong, dissipation goes up much faster than reduction of output. A scan down to 90% (just obvious) may be double the heat in the plate. So most TV Sweep ratings are very conservative to cover the difficulty of knowing actual losses.
Indeed. More generally, if you use the tap marked "x" watts for one anode, then the centre of that winding will be the tap marked "4 x", which goes to B+, and the tap marked "common" goes to the other anode.
If a tap is marked, say, 0.5W, you can work out the voltage needed to provide 0.5W to an 8 ohm load; this is the voltage across the transformer secondary.
Since we know the primary is being fed from a 70 V audio line, you can then calculate the transformer ratio, which is just the same as the voltage ratio.
Then you square the transformer ratios to get the impedance ratios.
Finally, multiply the impedance ratios by the speaker impedance (8 ohm in my case) to figure out the primary impedances.
I put together a spreadsheet to work out the numbers for a number of affordable 70V audio line transformers some years ago (pdf attached).
Sadly, none of the transformers I looked at provides a 6.6k end-to-end impedance at any tap. So none is really suited to George's 32ET5 recipe, unless you provide an oddball speaker impedance on the secondary side.
To use my PDF: let's say you pick an end-to-end primary impedance of 10,000 ohms. To find the centre-tap, find the tap with the impedance of one quarter of 10,000 ohms, i.e., 2500 ohms. That will be the centre-tap which goes to B+.
This will work out to be the same tap Printer2 mentioned; the centre-tap will be labeled with four times the power of the tap that provides the end-to-end impedance you want, and calculate out to one-quarter of the impedance.
On the plus side, the smaller the end-to-end impedance we need, the less of a problem the inductance is.
As an example, we only need one-fourth the inductance to provide the same bass response with a 4k transformer, as we do with a 16k transformer.
So this is another area where the 32ET5 and family might be ideal candidates, since they were designed to drive some pretty low-impedance loads (2500 ohms SE, etc). Some of the other low-power valves I looked at (like the 6AK6) really want 10,000 ohms single-ended and 20,000 ohms end-to-end in push pull.
-Gnobuddy
That's never going to get happy.
The transformer is already in distress with 70V of background music. With push-pull it can about take 35V each side. Peak 50V. Assuming a tube should pull-down 70% of supply (already poor efficiency) the supply should not be much over 72V. Exceed that and bass will come out mangled. So a small guitar amp may do OK with 70V iron and 100V supply, but we are far-far from George's 200V-400V supplies.
There are 100V transformers outside the US/CAN market. Gives a bit more.
There are 200V speaker transformers but they make sense more for KiloWatt ballpark systems than background music system, mostly hi-buck. "Such high voltage systems have been used in locations where small diameter wire is already in place, where long distance wire runs are involved and at especially loud installations such as Daytona International Speedway and the Indianapolis Motor Speedway prior to its redesign in 2003. Safety considerations involved with such high voltages require speaker line installation within conduit in most of the world." link
Last edited:
Something isn't right. There are lots of DIY guitar amps in Australia using 100V audio line out transformers and B+ in the typical 300 - 400V range, and lots more in North America using 70V audio line transformers.
I've seen proper output measurements made on some of these amps by techs who know what they're doing (complete with 'scope shots and meter measurements.) The amps and output transformers are not blowing up and failing everywhere you look.
Poke around the Aussie Guitar Gearheads (AGGH) forum, and the section on cheap transformers on the late Roly Roeper's website ( Cheap Output Transformers ) and see for yourself.
The Australian DIY guitar amp community seems to have pioneered this idea of using audio line transformers as valve output transformers, mainly because of the extremely high cost of "proper" valve output transformers there.
-Gnobuddy
We, or at least I know that the $4 transformer lives in a push pull 32ET5 amp making 4 watts from 170 volts B+.
So what would happen if I take the big OPT out of my test board and wire this little guy in its place and turn the power supply up to 400 volts?
Stay tuned. It's getting dark aand I need to grill myself some dinner before darkness falls. I will have a full report "complete with 'scope shots and meter measurements."
Attachments
My experiments for most of yesterday were done with 400 volts on 6AQ5's, so that's where I started. I thought that the worst that could happen was I would fry a $4 transformer. well, that didn't happen.
I tested the tiny transformer with 6AQ5's with an 8 ohm load on the 4 and 8 ohm taps at 1 KHz and at 82 Hz. I expected crappy bass performance, and I got it. The bass is there and doesn't sound as bad as it looks, but you can decide how much distortion you can live with and build your amp with that much power. The 10% choice is beyond the point where I would run the tubes.....limit the screen current to avoid meltdown. The current was about what I expected at of below 5% distortion......What I didn't expect was how well these guys passed power, and how good they sounded playing music.
The trick of putting the 8 ohm speaker on the 4 ohm tap doesn't work here. Saturation is primarily caused by AC voltage across the primary, and a 10K load puts about 600 volts P-P across the primary.....no fireworks, just uber distorted bass. Max power at 82Hz was under 3 watts and the tube current was over 200mA.....no good.
I put the 8 ohm load where it belonged and turned up the drive until each distortion level was reached, then recorded the output power. Yes, you can shove 37 watts through the tiny transformer without fireworks.
6AQ5 data for 8 ohm load on 8 ohm tap:
Distortion %: 1% 2% 3% 5% 10%
Power at 1 KHz W: 21W 29W 32W 34W 37 W
Power at 82 Hz W: 1.4W 4.4W 6.3W 9.7W 15 W
I unplugged the 6AQ5's and plugged in some 6GC5's (9 pin 6W6) readjusted the bias and cranked it up. It has been my experience that a bigger tube will improve the performance of a wimpy OPT for a given power level, or raise the power output for a given distortion level.....that applies here too.
6GC5 data for 8 ohm load on 8 ohm tap:
Distortion %: 1% 2% 3% 5% 10%
Power at 1 KHz W: 31W 34W 35W 35W 35 W Clipping reached @ 35 W
Power at 82 Hz W: 1.7W 4.6W 6.5W 11W 16.3 W
At this point I connected my thrift store Cerwin Vega speaker up and drove the amp with my iPAD. I played music and was surprised at the quality of the sound. For most of what I played there wasn't much difference between the tiny transformer and the big one sitting behind it. Even some bass heavy music came through the speaker with the same apparent bass that the big transformer had, but these speakers don't do big bass. They have a 4 inch "woofer." I took some pictures with the same music as yesterday, and they look similar. I can still see 45 to 50 volts P-P across the speaker.
I hooked up my blow proof test speakers (2 X 6 inch PA drivers + tweeter) and cranked some music. This amp gets quite loud through these speakers, and while it won't kick one of my other amps off the shelf, it might be something worth building......
So these little guys may be useful for a super low buck guitar amp in the 10 watt range. I need to try them with a real guitar speaker that has a resonant frequency in the 80 to 100 Hz range. The impedance peak may drive the tiny transformer nuts!
I'm waffling with the idea of simply wiring a 340 volt power supply up to the OPT CT in the 32ET5 amp that I have now to see what happens. It can't do AB2, so the power should be limited to maybe 10 watts......what have I got to lose, a $4 OPT?
Attachments
I tried it and my guess was correct 10 watts happen. That amp is cathode biased. Simply stuffing 340 volts on the plates makes the idle current 50 mA per tube so the cathode resistor must go from 150 ohms to 250 ohms. That gets the idle current down to 25 ma per tube and burns 20 volts in the resistor. 25 ma is still a bit high so more resistor, and more wasted voltage is needed.
Going from 4 watts to 9 or 10 watts is not worth hacking up my working amp, so some more experiments are in order.
I'll have to try my 12AQ5 amp on higher voltage, I have them going into a 70V transformer. I think they have 260V on them and they sound kind of wimpy. I have a switch to change the output tap from 8 to 4 ohms and as you I never notice much difference. If I got in the vicinity of 10W I might rebuild it and give it to my nephew.
Thank you for your report(s)! But you aren't just a tech, you have a couple of engineering degrees under your belt...that's a whole different kettle of fish!
(Of course there are plenty of techs who do have a solid understanding of what they're doing, but every once in a while you hear one babbling on about guitar harmonics dying in a semiconductor crystal, or something else equally ignorant...)
Now that is very good news to me! A 32V heater requirement makes it a bit tricky to provide heater power to the preamp valves (though one could string 5 identical 6.3V preamp triode heaters in series for 31.5V). But the 50C5 comes in a whole family of different heater voltages.
So I looked in the junk box, err, treasure chest, and I have four 17C5s and four 25C5s.
One can string two 12AX7 heaters in series and power them off 25V. I have a couple of 24V DC switching power supplies which should be able to power two 25C5 heaters (in parallel) and the two 12AX7 heaters (in series). Four triodes and two output beam tetrodes - those are the basic ingredients for any number of small Fendery amps.
It also occurs to me that three 6.3V heaters (6AQ6 triodes, say) in series adds up to 18.9 volts, and most laptop power supplies are rated at 18V or 19V. A pair of parallel 17C5 heaters should be happy on a laptop power supply if you drop a volt or two in a small-value series 5 watt resistor of around 1 or 2 ohms. And I also have a couple of old laptop power supplies in the treasure chest.
A slightly more awkward part of the equation will be deciding how to providing positive and negative rails for the driver MOSFETs. If I end up using DC to power the heaters, it might make sense to also use that same DC rail for one of the two MOSFET power rails. I just have to find a way to generate the other polarity MOSFET power rail then.
Couple of minor typos there, it should be 0.625W rather than 0.0625W, and the 70V/2.5W tap is the centre-tap that goes to B+, not the other end. (The anodes go to the "common" and 0.625W taps.)
I know George and Printer2 know what they're doing, but just in case someone else reading this thread gets confused, I thought it worth making the correction.
My spreadsheet (which assumes ideal numbers) predicts a 31.3:1 turns ratio for the 0.625W tap of a 70V transformer, and this results in 7840 ohm primary impedance.
Meantime, my calculator thinks a 35:1 ratio with an 8 ohm speaker works out to 9800 ohms.
I remember reading somewhere that "70 V" transformers are actually rated for 70.7 volts nominal, which might change the numbers very slightly. But I doubt very much these transformers are wound to that sort of (1%!) accuracy.
And there's the bottom line, whatever the actual Raa of the $4 transformer might be!
I thought for a moment that it might be possible to double the power handling, as well as primary and secondary winding inductances, by series-connecting the primaries of two $4 transformers, and also series-connecting the two secondaries. The junction of the two primaries becomes the centre-tap.
Everything seems to work out for AC currents. But there is a fly in the ointment: DC magnetization from the quiescent current in each output valve does not cancel, because each valve's idle current only flows through one transformer primary, and therefore only magnetizes that one transformer's core.
However: George's high-power recipe does include very low quiescent anode currents to keep the power dissipation reasonable with the high supply voltage. So maybe this will work?
If the $4 transformer can cope with the rather small 12 - 15 mA of DC current through the primary, then becoming a big spender and shelling out $8 for two transformers might work (attached image).
I note that applying a 330V peak (AC) half sinewave to a 3300 ohm (half-primary) results in 100 mA peak AC current from each anode, which the $4 transformer apparently copes with (17W RMS, well within some of George's measured numbers.)
So if the transformer can cope with 100 mA AC peaks, I would think it could handle a 12-15 mA DC current, which is 7 or 8 times smaller. (This is very different from the class A, single-ended situation, where idle current is as large as peak AC current, doubling the total current through the transformer on negative-going peaks.)
There's only one way to find out, I suppose!
-Gnobuddy
Attachments
Last edited:
I'm now wondering what voltage the insulation on these little transformers is rated for.
In an ideal world where the output valves could pull their anodes down to zero volts on negative peaks, the peak-to-peak voltage across both ends of the primary winding would go up to four times the DC supply voltage.
Realistically, a 300V B+ and, say, 50V saturation still gets you 1000 volts peak-to-peak across the two ends of the "70 V" audio line transformer.
That is worryingly high, and I wonder if there is not a risk of flashover, either between the two ends of the primary, or worse, between primary and secondary windings.
Perhaps it would be wise never to touch the speaker terminals when using one of these little audio line transformers as valve output transformers. That really means combo amps only.
-Gnobuddy
I got the engineering degrees at my company's strong suggestion (and their expense) after working my way up from assembly line tech to engineer by making and fixing stuff that the degreed boys couldn't or wouldn't do.
I have concentrated on connecting tubes of like heater CURRENT in series across the isolation transformer's output. There are other ways.
My little amp uses an 18FW6 (pentode like a 6AU6) and 18FY6 (triode like 1/2 a 12AX7). These eat 18 volts @ 100 mA and were designed for the last generation of tube radios as was the 32ET5. Feed them directly from the laptop brick. They are plentiful for now and $1 each.
I remember seeing some positive to negative voltage converter circuits in the applications notes of a switcher chip. I don't remember which one though.
Last night I thought of an old idea that I have visited several times with differing results. There has been some discussion of a high output power mode created by driving both G1 and G2 at the same time on this forum for over 10 years. There was an old schematic of a PA amplifier that extracted over 100 watts from a pair of 807's (6L6GA's). I remarked that saw a publication from a tube vendor showing high power from a pair of common tubes, but I could not find it. I built a simple test circuit in a forum post about 8 years ago. It is in post #30 here:
G1=G2/mu Scaled Drive Strawman Design
The tube publication is in the GEC KT88 application notes. They get 150 watts from a pair of KT88's on 750 to 850 volts. The app note is too big to attach (9MB), A screen grab of the schematic is included. No negative voltages are needed for this or pure screen drive, both will be tested.
I obviously get confused while walking from the bench to the PC.
A 35:1 ratio into 8 ohms is 9800 ohms, that's why the tubes didn't like it and the transformer saturated.
I simply connected my audio oscillator and audio analyzer in parallel across the whole primary and adjusted for 5 volts RMS, then moved the analyzer to the secondaries and recorded the voltage output. I got .143V on the 4 ohm output and .202V on the 8 ohm output. That's a 35 to 1 on the 4 ohm tap, and 25 to 1 on the 8 ohm tap.
That's the scary part. I don't see much if any tape between the primary and secondary. My experiments always have the secondary grounded to avoid blowing up my test equipment when something goes wrong. So do my amplifiers. So far no fireworks....on ONE test sample. The next transformer could fry!
Attachments
What do you need for negative voltage and current?
DC-DC 3.6v-36v MC34063A Positive Voltage Negative Voltage Module K9 | eBay
The reverse voltage module MC34063A_NV is a chip based on MC34063A for providing reverse voltage, negative voltage modules. Capable of converting + 5V voltage to -5V voltage output. The output voltage can be adjusted by adjusting the potentiometer according to the requirements.
Technical Parameters:
Input voltage range: 3.6V-36V.
Output voltage range: negative 1.5V to negative 36V. Output voltage regulation characteristics; load regulation rate of less than 5%.
Output current. The IC of the module is 1.5A internal switch, but for safety and heat considerations, it is recommended that the load power of this module is less than 0.5W. That is, the output negative 5V, the load within 100MA, the proposed 50MA.
Other parameters:
Using single-chip embedded technology.
Input positive voltage, output negative voltage.
Easy to operate, you can adjust the output voltage through the potentiometer.
The module is small, easy to carry and use a variety of occasions.
Note: the higher the negative pressure, the load capacity will be relatively down.
Boost Converter 3-18V to Positive Negative Step-Up Power Supply Module +/-24V | eBay
1.Input Voltage: 3.6-18V
2.Output Voltage: +/-24V
3.Max Input Current: 1.8A
4.Max Output Current: +Vo 400mA; -Vo 100mA
5.Quiescent Operating Current: 3-4mA
6.Operating Frequency: 400KHz
7.Operating Temperature: -40~125 Celsius
8.Storage Temperature: -65~150 Celsius
9.Size: 25*16.3*6.6mm
DC-DC 3.6v-36v MC34063A Positive Voltage Negative Voltage Module K9 | eBay
The reverse voltage module MC34063A_NV is a chip based on MC34063A for providing reverse voltage, negative voltage modules. Capable of converting + 5V voltage to -5V voltage output. The output voltage can be adjusted by adjusting the potentiometer according to the requirements.
Technical Parameters:
Input voltage range: 3.6V-36V.
Output voltage range: negative 1.5V to negative 36V. Output voltage regulation characteristics; load regulation rate of less than 5%.
Output current. The IC of the module is 1.5A internal switch, but for safety and heat considerations, it is recommended that the load power of this module is less than 0.5W. That is, the output negative 5V, the load within 100MA, the proposed 50MA.
Other parameters:
Using single-chip embedded technology.
Input positive voltage, output negative voltage.
Easy to operate, you can adjust the output voltage through the potentiometer.
The module is small, easy to carry and use a variety of occasions.
Note: the higher the negative pressure, the load capacity will be relatively down.
Boost Converter 3-18V to Positive Negative Step-Up Power Supply Module +/-24V | eBay
1.Input Voltage: 3.6-18V
2.Output Voltage: +/-24V
3.Max Input Current: 1.8A
4.Max Output Current: +Vo 400mA; -Vo 100mA
5.Quiescent Operating Current: 3-4mA
6.Operating Frequency: 400KHz
7.Operating Temperature: -40~125 Celsius
8.Storage Temperature: -65~150 Celsius
9.Size: 25*16.3*6.6mm
Last edited:
You need enough voltage to completely cut the output tube off when it's plate is at twice the supply voltage. In HiFi applications I aim for twice that much. The current demands are pretty low I use 10 mA idle current through each mosfet in a HiFi amp, but it could be less than half that.
The chip number used in that module rang a bell and sure enough that's the one I had used years ago.....long enough that my datasheet still says Motorola on it. The positive to negative converter circuit is on page 8 of the current On Semiconductor datasheet. It's good for 35 to 40 volts output and at least 50 mA. good enough for this job.
I spent the afternoon tinkering with a no-negative voltage circuit similar in concept to what I discussed earlier. Drive is directly fed to G2, and restively divided down to G1. I could get 35 watts out of a 6GC5 on 400 volts with good distortion at all power levels. The 6GC5 should do better on 400 volts so I tried a 50B5. This little guy screamed, but I could not get the distortion good at all power levels. After losing my patience and turning the knobs to the right, I found the red plate point on a 50B5. I was squeezing over 50 very distorted watts out of a pair on 450 volts while poking around with a scope probe. I realized that the distortion issues were because the driver board was running out of grunt. I needed more than 250 volts P-P from a 450 volt supply. There are ways to do this, but not with this board.
After poking around with a scope for about half an hour I noticed a faint red glow on the back side of one tube. These are no name 50B5's, probably Japanese, but they could be RCA's. Who puts 450 volts on a 50C5 anyway?
Attachments
Most of the degreed electronics engineers I knew didn't like to actually build anything. They would draw up a schematic on their computer, then hand it to a lowly underpaid tech to build.
Apparently most of them didn't care about the physical implementation, and figured the job was done once the schematic was produced and the tech brought them some test numbers. I never really understood this approach.
Excellent idea! Unfortunately for me, I scoured the ESRC dollar list for an hour yesterday, looking up data sheets for each unfamiliar type designation, and I found the 6AQ6, but not the 18FY6. I ended up ordering a few 6AQ6 and 12C5s.
Silly me, I looked up dozens of 4V and 5V and 6V and 12V parts, but not the 18V ones!
I've seen a few ready-made polarity reversing DC-DC converters like the one Printer2 linked to. I suppose there is also the option of simply grounding the positive lead of a regular switching wall-wart (after making sure there is DC isolation wrt the mains ground.)
I've read a few of those old threads, which often seemed to end with exploded MOSFETS.
I remember one of your posts where you described a catastrophic failure mode for overdriven guitar amps with screen drive: the screen grid would overheat enough to start emitting electrons itself, causing explosive runaway overheating and valve failure.
Since the only interest I have in valves is for use with electric guitars, I've had no desire to pursue screen drive experiments myself.
But getting 15 watts (never mind 35!) out of a pair of $1 seven-pin radio valves - that's hit a very sweet spot in my world. Fender Princeton power levels, without the $70 cost of a pair of current-production 6V6 valves.
Thanks for clearing that up! Okay, so 5k it is. (There was a typo in your earlier post that mentioned a 35:1 ratio for both 8 ohm and 4 ohm taps, and I made the wrong guess while trying to figure that out.)
What do you think of the "two $4 transformers in series" idea? If it works, it also halves the peak voltage across each transformer winding. Not a huge reduction, but at least a step in the right direction.
If that doesn't work, I think I have a nominally 15 W, 8 k:8 ohm Hammond guitar output transformer in the junk box. If an 8 k primary is too hard on the screen grids, I'll either increase the screen grid resistors, or put a 39 ohm power resistor in parallel with the 8 ohm speaker to drop it to 6.6 ohms.
-Gnobuddy
The 50C5 datasheets show around 15 mA of anode current with 250V on the anodes, 110V on the screen grid, and (-14V) on the control grid.
Since we're aiming for an idle current below 15 mA, and expect to have 340V on the anodes, we need more than 14 volts negative. Maybe a (-25 V) rail would do?
-Gnobuddy
I saw the handwriting on the wall maybe a dozen years ago. All the older people were being laid off, or shoved into crappy jobs until they left. I went to the boss and asked what stuff he needed done that the "simulating, calculating boys" didn't want to do. He said he needed someone to design and build prototypes and EVB's (evaluation boards) that proved our stuff actually worked. Our group had evolved into an IC design center, which is not a job I wanted. OK, let me get this right, you want me to MAKE stuff and I have a mechanical design center, a parts house, a PCB shop and an SMD assembly line IN HOUSE at my disposal......uh, YES. About 4 years ago that all died, I got a letter from the CEO offering me a lot of money to leave quietly. Recon informed me that all the support groups were going to be shut down and all the people laid off. That turned out to be true except for the SMD line.....I left.
It's gotten pretty bad lately. My fingers have become permanently numb and tingly. This afternoon I smelled my finger burning before I felt it when I tried to solder myself into a circuit. Playing guitar has become impossible and I have to proof read everything I type. Even in the last year at Motorola I had a tech solder the small SMD stuff for me.
Yeah but I was extracting well over 100 watts from a tube that is STILL on the dollar list in 12, 17 and 25 volt flavors when I did it. Those tubes will live forever at 50 watts. The GA version lives at 80 watts in a guitar amp.
That one is a 6AT6 with a 150 mA heater, useful for battery operated stuff as is the 6AK6. I have a bag full of them.
I got a pair of the 12 volt to 380 volt DC converters on Amazon. It took me 2 minutes to figure out that they will not live on a 19 volt laptop brick. I also got a 12 volt 5 amp brick from Amazon. I haven's blown it up yet.
I tried it with two of the big transformers that I use on my breadboard. It doesn't work. The DC current doesn't cancel and distortion reigns supreme. Wiring them in parallel does work if you halve the impedances one both ends. Two 6600 ohm to 8 ohm OPT's in parallel makes a 3300 ohm to 4 ohm OPT at almost twice the power. You can also use 4 tubes, a pair to drive each OPT, then series or parallel the secondaries. I have even done 8 tubes, 4 OPT's with their secondaries wired in series - parallel to make 300 watts from cheap OPT's.
There are plenty of tubes on that list that will make 15 watts per pair, without bending any specs. The 6EZ5 was a drop in in most 6V6 circuits and was on the list until someone here talked them up, then poof they were all gone. There is still a 12 volt octal and a 9 pin mini, both like the 6W6 that will make 15 watts per pair.
It's really sad what business as usual has turned into.
I watched a documentary recently ("Dirty Money" on Netflix) about a pharmaceutical company whose CEO boasted that research and development was a waste of time and money; his approach was to continuously acquire other pharmaceutical companies with existing products, then raise the price of those products by up to a thousand times.
I'm very sorry to hear about the health issues you're dealing with. Sometimes it feels as though growing old is just a series of progressively more extreme indignities that nature forces down one's throat.
I have one of those - the ad claimed it would indeed live on 19V, and that had been my plan. So now we know - the hard way - that it won't. Great.
Good to know, but unfortunately, that reduces their effective primary inductance, too. Too bad the series connection doesn't work.
Inspired by this thread (and an upcoming birthday), I did order a few different types from ESRC, including a few of a 12-watt, $3 valve.
The 50C5 and family are interesting in part because they have been treated with so much scorn, and now turn out to be vastly more capable than anyone had realized in the last fifty years.
The 50C5 curves also show an enormous amount of nonlinearity in curve spacing, so I have a feeling they might make really nice guitar "clean tones" in single-ended mode.
I am curious what happens with a pair in push-pull if we add a little local negative feedback around only one of the pair to linearize it a bit, and then drive it a bit harder to maintain balance with the lowered gain. That should substantially reduce cancellation of even harmonics. SE tone from a pp amp, or just unpleasant crossover distortion?
-Gnobuddy
It occurred to me belatedly that dropping from 37W to 15W is only a 4 dB drop; normal power bandwidth is defined between the -3 dB or half-power frequencies.
So the tiny $4, ten-watt transformer with four times its nominal power pushed through it at 1 khz, still actually managed to work down to 82 Hz only 4 dB down from 1 kHz. I would call that a rather spectacular success.
I can see why there would be little to no audible bass shortage with guitar. Most guitar speakers don't go down to 82 Hz with less than 3 dB drop-off, either.
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.