Hello,
I am starting this project. I already have the Allied 6K7VG Transformer, but haven't ordered the opt's yet. My intention is to build it as simply as possible for now. I'll temporarily leave out the choke and cap. I'll also stick with ss rectification for now. (those items are likely to change in the future) My goal is the most power I can get because of my large inefficient 4ohm speakers. I'll use UL and CFB. I can and probably will stick with kt88's. My real question is, should I go with a lower impedance opt and eliminate the ability to use el34's opting for a couple more watts on the kt88's? Or, is the power gain insignificant and should I just stick with the 5k so I have all the tube options? Here are my tranny choices. All Edcor because of the value;
GXSE15-4-3.5K for strict kt88 use and budget consciousness.(I'll need help on resistor selection)
GXSE15-4-5K for use of all tubes 34/88/6l6 and budget consciousness.
CXSE25-4-2.5K all out balls to wall kt88 only(with resistor help from you guys)
CXSE25-4-5K This will be a no compromise/build exactly as laid out on tubelab sight.
Please help because I want to get started.
I am starting this project. I already have the Allied 6K7VG Transformer, but haven't ordered the opt's yet. My intention is to build it as simply as possible for now. I'll temporarily leave out the choke and cap. I'll also stick with ss rectification for now. (those items are likely to change in the future) My goal is the most power I can get because of my large inefficient 4ohm speakers. I'll use UL and CFB. I can and probably will stick with kt88's. My real question is, should I go with a lower impedance opt and eliminate the ability to use el34's opting for a couple more watts on the kt88's? Or, is the power gain insignificant and should I just stick with the 5k so I have all the tube options? Here are my tranny choices. All Edcor because of the value;
GXSE15-4-3.5K for strict kt88 use and budget consciousness.(I'll need help on resistor selection)
GXSE15-4-5K for use of all tubes 34/88/6l6 and budget consciousness.
CXSE25-4-2.5K all out balls to wall kt88 only(with resistor help from you guys)
CXSE25-4-5K This will be a no compromise/build exactly as laid out on tubelab sight.
Please help because I want to get started.
Thanks. The thing is, with the 15w transformer, will it be ok with the kt88's in ul with cfb and ss rectification? Do you know the wattage that combo will produce using the 560r or 510r? If that transformer is safe I wouldn't mind saving the money. My speakers don't do much below 40hz anyway.
If you need to drive full range, I would spring for the CXSE's. As to the primary impedance, that depends on your speaker. You say it's 4 ohms...how flat is the impedance curve? You might be better off with the better DF of the 5k primary over the couple of watts of power. boywonder also has a good idea.
How would that effect the bass response? Usually using 8ohm tap with 4ohm speakers causes you to lose bass.
'Lose bass' can have different meanings - the usual expectation would be to have a bit more power and distortion and perhaps more, but 'looser' bass. Any time I experimented with this, I found the difference was less dramatic than I expected.
This may have to do with the impedance vs freq characteristic of the speakers being used, of course.
It would be handy (though more expensive, I guess) if Edcor offered multi-output taps as standard.
VictoriaGuy has a good point; reducing the effective output transformer primary impedance (ie connecting a 4 ohm speaker to an 8 ohm tap) usually reduces the damping factor and makes the bass "looser" or "flabby". This is sometimes perceived as more bass. I notice this effect quite a bit when I run my full range speakers (half changs) on a SS amp the bass is very thin to non-existent compared to using my TubeLab SE. The SS amp has a much higher damping factor than the Tubelab SE. Conversely, some commercially designed speakers will sound bass heavy when driven by a no feedback, SE triode amp.
I tend to try to focus less on the claimed wattage rating of the output transformer, and instead seek to learn the actual maximum current rating of the primary winding. Especially with single ended transformers, you need to make sure you aren't exceeding the max current spec, otherwise you'll drive the core into saturation. You can most likely assume your KT88 tube will idle somewhere around 70 mA or less. Even if the 6K7VG can keep the B+ up to 450VDC under that kind of draw (optimistic) you're talking about 30+ watts of idle plate dissipation per tube. That's plenty enough for KT88, especially modern production. So the only question that remains is this - is the GXSE15 sufficiently gapped for 70mA standing current? Phyllis might be able to answer that.
I'd wager the answer is yes.
I have an SSE that uses EH KT88's, Transcendar OPT's and is powered by a 6K7VG. I usually run it with the tube rectifier and the B+ is about 440 volts if I am remembering right. I made some measurements and found that I get the lowest distortion a n best sounding bass when I run the tubes at 100 mA each. With the solid state rectifiers enabled the B+ is over 450 volts. This much current is not recommended since it is way over the max spec for the Hammond build Allied 6K7VG. I have been running it this way off and on for about 5 years and the transformer is still going strong, although it gets too hot to touch at this current level.
The OPT's in that amp are designed for 300B tubes and are 3000 ohms with a 4 and an 8 ohm tap. The EH KT88's do not like 3000 ohms into my speakers. I get noticible distortion on bass peaks, so I run my 8 ohm speakers on the 4 ohm tap for a 6000 ohm load. I can actually play it louder since it doesn't distort. The speakers in question are above 8 ohms for most of the audio range and are well above 10 ohms below 100 Hz. If your speakers need good damping I would go with the 5K OPT's.
I have no experience with the GXE series Edcors but I can report that the cheaper XSE-15 series works just fine at 80 mA with 300B tubes. Those transformers will not work below 100Hz at 15 watts. I doubt that the GXE's will get down to 40Hz at 15 watts. The CXSE-25's will go to 25Hz at 15 watts.
Thanks for setting my memory straight. I thought the 6K7VG would sag quite a bit under that much load. Apparently I was mistaken.
I forgot to mention my 122 volt line voltage. You might get a little less on a more normal wall socket. I am 1200 miles from my lab right now so my numbers could be off a bit, but I know it had plenty of power even heavilly overloaded.
Also note that Hammond and Allied transformers have changed a bit over the years. I get more voltage out of a new one than a 5 year old one in the same amp on the same test bench. The sag performance may differ too. The transformer I put in that SSE was a few years old when I built the amp, and the amp is about 5 years old.
Quite a while ago (7, maybe 8 years) Allied gave me a 10% off coupon and offered free shipping on orders over $100. That was before the price increases too. I got about 100 pounds of transformers on one order. I still have a few hanging around.
Also note that Hammond and Allied transformers have changed a bit over the years. I get more voltage out of a new one than a 5 year old one in the same amp on the same test bench. The sag performance may differ too. The transformer I put in that SSE was a few years old when I built the amp, and the amp is about 5 years old.
Quite a while ago (7, maybe 8 years) Allied gave me a 10% off coupon and offered free shipping on orders over $100. That was before the price increases too. I got about 100 pounds of transformers on one order. I still have a few hanging around.
I have bought a 193H choke with 65ohm DCR. Is it fine to use by itself or would I need more series resistance as suggested below. Also, if additional resistance is required, what would be the reason for it? Is it to drop B+ further as 65ohm wouldn't drop it to within a safe range?
It will work, but some additional series resistance will improve filtering. In a CRC filter, which is the original design, increasing R will improve the filtering of the second C. Of course, the downside is a lower B+. A CLC filter gives you even better filtering, because the inductor stores energy along with the capacitors. 5H is not really enough inductance on it's own, but in combination of the higher DCR of the Allied choke it works out.
It doesn't hurt to try it. If you get more hum than you'd like, add more series resistance or increase the size of the second cap, The resistance also helps isolate the second capacitor from the rectifier. Since there is a choke there I don't think it is cause for concern.
It doesn't hurt to try it. If you get more hum than you'd like, add more series resistance or increase the size of the second cap, The resistance also helps isolate the second capacitor from the rectifier. Since there is a choke there I don't think it is cause for concern.
Could I please confirm the latest suggestions on how to connect the amp compared to the schematic. I understand that the standby switch should not be used as it can generate a voltage spike, so CT should be permanently connected to amp ground? Also in regards to the inrush limiter, I couldn't find a diagram in previous posts so could someone confirm I have the connections right for 374BX and 220V operation. If I am correct, the limiter should go in series with the transformer primary and also in between the CT and ground? Lastly, the thermistor I can easily buy is a 22ohm 2.8A and not the 120ohm recommended, that should be fine? Should the thermistor connected to the CT be of lower resistance to the one connected to the primary?
Attachments
IRCL devices are specified by the cold resistance, hot resistance, and maximum current they can handle. For our type of applications, we generally want as high a cold resistance and as low a hot resistance as possible. The maximum current must be a relatively close match for the expected current draw.
The CL-90 is particularly well suited for the primary side of many vacuum tube power amplifiers. The cold resistance of 120 ohms is one of the highest values you will find among the CL type devices. At 50% of full rated load, the CL-90 has a hot resistance of three ohms, so it's getting close to being out of the circuit at that point. Finally, the 2 amp maximum rating fits well with the expected primary side current draw of many vacuum tube amplifiers.
For example, the Tubelab Simple Single Ended fitted with KT88 may require 170 mA of B+ at 450VDC, plus 3.5 amps of heater current at 6.3 volts and 2 amps at 5 volts. That's a total of 110 VA. Let's assume our power transformer is roughly 80% efficient (a number pulled completely out my back side), so expect a primary side current draw of 1.1 amps. That's well within the capabilities of the CL-90, but still close enough to its max load rating that the device will get hot.
I do not understand why people want to put IRCL devices on the B+ loop. Your current draw here will usually be less than 200 milliamps. A device like the CL-90 is, for all intents and purposes, still "cold" at this current draw. You may as well just stick a 120 ohm resistor between center tap and ground. Even something like a CL-140 (rated 1.1 amps max, 50 ohms cold, 5 ohms at 25%) is a stretch. IRCL devices do not function without heat, and to get them hot you need to have substantial current draw through them.
Just my two cents.
Oh, and yes you do have the CL-90 on the primary side in the correct location, although with your higher line voltage (220V) you can expect current draws of roughly half what we get in the States. You will probably be operating a closer to 25% of the CL-90 rated max load, where it has a resistance of nearly 8 ohms. Study the data sheets, and you may find an IRCL that is better suited for your application.
The CL-90 is particularly well suited for the primary side of many vacuum tube power amplifiers. The cold resistance of 120 ohms is one of the highest values you will find among the CL type devices. At 50% of full rated load, the CL-90 has a hot resistance of three ohms, so it's getting close to being out of the circuit at that point. Finally, the 2 amp maximum rating fits well with the expected primary side current draw of many vacuum tube amplifiers.
For example, the Tubelab Simple Single Ended fitted with KT88 may require 170 mA of B+ at 450VDC, plus 3.5 amps of heater current at 6.3 volts and 2 amps at 5 volts. That's a total of 110 VA. Let's assume our power transformer is roughly 80% efficient (a number pulled completely out my back side), so expect a primary side current draw of 1.1 amps. That's well within the capabilities of the CL-90, but still close enough to its max load rating that the device will get hot.
I do not understand why people want to put IRCL devices on the B+ loop. Your current draw here will usually be less than 200 milliamps. A device like the CL-90 is, for all intents and purposes, still "cold" at this current draw. You may as well just stick a 120 ohm resistor between center tap and ground. Even something like a CL-140 (rated 1.1 amps max, 50 ohms cold, 5 ohms at 25%) is a stretch. IRCL devices do not function without heat, and to get them hot you need to have substantial current draw through them.
Just my two cents.
Oh, and yes you do have the CL-90 on the primary side in the correct location, although with your higher line voltage (220V) you can expect current draws of roughly half what we get in the States. You will probably be operating a closer to 25% of the CL-90 rated max load, where it has a resistance of nearly 8 ohms. Study the data sheets, and you may find an IRCL that is better suited for your application.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.