Alright, but It is really really expensive to buy those (hi-fi) transformers, almost the cost of building another amp. and before I make any $150 decisions, I am going to get the best out of what I have then re-evaluate. I think I am going to put on the CFD right after I have posted this messege, and give it a listen.
I do use a pair of JJ El34s, and how much power do you think I could pull out of them? because I want to get 15WPC out of it (my friend's EL84 PP amps power), if at all possible to compare the two amps, I know this would have to occur on an UL basis. my Power Transformer is capable of pumping out 200ma, so assuming I run each tube a 90ma, even at 300v thats 27W of dissipation, which is way over spec.
In addition, what Tube do you think will get the most output power (I am 17....) I might just invest in a pair of KT-88s by JJ because they seem like workhorse tubes. but I wanted to see which tube is the best. now that this amp works pretty well, I am not afraid to invest a good 50-60 bucks on good tubes
I do use a pair of JJ El34s, and how much power do you think I could pull out of them? because I want to get 15WPC out of it (my friend's EL84 PP amps power), if at all possible to compare the two amps, I know this would have to occur on an UL basis. my Power Transformer is capable of pumping out 200ma, so assuming I run each tube a 90ma, even at 300v thats 27W of dissipation, which is way over spec.
In addition, what Tube do you think will get the most output power (I am 17....) I might just invest in a pair of KT-88s by JJ because they seem like workhorse tubes. but I wanted to see which tube is the best. now that this amp works pretty well, I am not afraid to invest a good 50-60 bucks on good tubes
15 WPC from a SE amp isn't going to be easy, or cheap. First of all I have hammered the 125CSE's pretty hard, and they will pass 15 watts at 1 KHz, but saturation is visible for anything above 5 watts below 100Hz.
I think that you are already near the limit for the 125CSE, especially in the bass department. Replacing the EL-34's with KT-88's or 6550's would only gain you a small amount of power in triode mode (a watt maybe). The only KT-88's that I have are older Chinese ones. I have already blown two of these up at power levels less than what I am currently feeding to the 6L6GC's. I have old version Sovtek and GE 6550's, these will work at 30 watts without any problem. I don't push them any harder because they are rare and expensive ($50 each for the GE's). How much do I get from them? 10 to 12 watts in UL depending on the transformer. Keep in mind that I am running 475 volts of B+ and a 5K load, so your results will be different.
I have not tried any JJ tubes other than the EL-34's. I bought several different types of EL-34's and the JJ's sound the best, and seem to handle some abuse, but I have not used them for long.
So, if you had the ultimate transformer, none of the popular tubes can RELIABLY pump out 15 watts. The usual choices for a 15 watt SE amp are transmitting tubes which require a lot of voltage, or a pair of the popular tubes wired in parallel for each channel. This is the easiest, and safest approach to take. Yes, I am getting 15 watts out of my severely overloaded 6L6GC's in UL mode, however I wouldn't be surprised if one of them exploded tonight. I am purposely stressing this amp to find any weaknesses in the design. I don't mind risking $4 tubes in the name of science.
If It were my decision on a limited budget, I would save the cash for upgrading the transformers later. The transformers are more responsible for an amps sound than anything else.
I will be doing some parallel tube testing in search for more power at a later date. I am planning on building a guitar amp using both channels of a SimpleSE in parallel.
Congradulations on building a sucessful tube amp at age 17. I know that you will be making more, this is an addictive hobby. I began at a young age by building guitar amps and selling them to my friends. Back then tubes, transformers, and all of the other electronic components were FREE, at the local trash dump (1960's).
I think that you are already near the limit for the 125CSE, especially in the bass department. Replacing the EL-34's with KT-88's or 6550's would only gain you a small amount of power in triode mode (a watt maybe). The only KT-88's that I have are older Chinese ones. I have already blown two of these up at power levels less than what I am currently feeding to the 6L6GC's. I have old version Sovtek and GE 6550's, these will work at 30 watts without any problem. I don't push them any harder because they are rare and expensive ($50 each for the GE's). How much do I get from them? 10 to 12 watts in UL depending on the transformer. Keep in mind that I am running 475 volts of B+ and a 5K load, so your results will be different.
I have not tried any JJ tubes other than the EL-34's. I bought several different types of EL-34's and the JJ's sound the best, and seem to handle some abuse, but I have not used them for long.
So, if you had the ultimate transformer, none of the popular tubes can RELIABLY pump out 15 watts. The usual choices for a 15 watt SE amp are transmitting tubes which require a lot of voltage, or a pair of the popular tubes wired in parallel for each channel. This is the easiest, and safest approach to take. Yes, I am getting 15 watts out of my severely overloaded 6L6GC's in UL mode, however I wouldn't be surprised if one of them exploded tonight. I am purposely stressing this amp to find any weaknesses in the design. I don't mind risking $4 tubes in the name of science.
If It were my decision on a limited budget, I would save the cash for upgrading the transformers later. The transformers are more responsible for an amps sound than anything else.
I will be doing some parallel tube testing in search for more power at a later date. I am planning on building a guitar amp using both channels of a SimpleSE in parallel.
Congradulations on building a sucessful tube amp at age 17. I know that you will be making more, this is an addictive hobby. I began at a young age by building guitar amps and selling them to my friends. Back then tubes, transformers, and all of the other electronic components were FREE, at the local trash dump (1960's).
Thank you very much! I already want to build an EL34 PP design that should yeild 15-18wpc in triode mode, but the preformance of my current amp may make me reconsider, the truth is I am pretty content with the power I can pull out of this amplifier.
I think the next inevitable phase for this amplifier is to replace an RC filter in the PSU with an LC filter, I have one choke already, but I want the lower ripple, and the raise in B+ to something a little higher, to increase versitility.
I may just order some EDCORs so I can run UL operation, some of the EL cheapo ones, but I want to get my primary OPT resistance set before I invest anything
I think the next inevitable phase for this amplifier is to replace an RC filter in the PSU with an LC filter, I have one choke already, but I want the lower ripple, and the raise in B+ to something a little higher, to increase versitility.
I may just order some EDCORs so I can run UL operation, some of the EL cheapo ones, but I want to get my primary OPT resistance set before I invest anything
Hey, late last night, I put the CFB on, it was amazing, the Bass was so much tighter, it made everything sound a little more relaxes, more smooth.
However my non-audiophile (senior year physics major) brother told me that he thought the upper, higher ranges sounded a little more muffled than before, I personally couldn't tell the difference. However is it possible that the sheer cheapness of my Cathode bypass capacitor causes damage to the higher frequencies? what would be a suitable replacement?
Also, the addition of CFB did lower the volume a little bit, however it seemed to push back the clipping point (the physical place on the volume nob in which clipping begins)
CFB was WELL worth it, and I plan to leave it on.
However my non-audiophile (senior year physics major) brother told me that he thought the upper, higher ranges sounded a little more muffled than before, I personally couldn't tell the difference. However is it possible that the sheer cheapness of my Cathode bypass capacitor causes damage to the higher frequencies? what would be a suitable replacement?
Also, the addition of CFB did lower the volume a little bit, however it seemed to push back the clipping point (the physical place on the volume nob in which clipping begins)
CFB was WELL worth it, and I plan to leave it on.
Hi Alex,
Try bypassing the cathode bypass caps with small films, say several uF in parallel and see what that does for you..
One option would be to replace them with a couple of ASC PPIO cans which you can usually find on eBay. These generally are not available in values much over 100uF though, which is still probably ok for what you are trying to do.
Another option would be to invest in a pair of rubycon blackgate electrolytics which I have found to sound good in these sorts of applications - the value and voltage required would make them expensive though. (You would need a 50V cap minimum.)
Still another option if you are up for a challenge would be to switch to fixed bias operation. I don't know whether or not your transformer has a bias tap or not, but at worse you could add a small transformer to get the voltage required. Typically an EL34 (depending on brand) might require somewhere between -28V and - 36V on the grid to achieve about 60mA of idle current at around 350V - 400Vdc on the plate. Fixed bias would give you the option to tweak the operating point for your tube/transformer combo and eliminate the cathode cap altogether. Typically you would use about a 10 ohm resistor in series with the cathode connection and measure across it check and set the operating point.
In terms of A/B tests between these two amplifiers, as long as you operate both at levels well below the clipping threshold of yours you'll be ok. The most important thing of all is to make absolutely certain that the levels are closely matched. Two amplifiers of otherwise equal virtue but differing loudness will unfairly give the win to the louder of the two.
Try bypassing the cathode bypass caps with small films, say several uF in parallel and see what that does for you..
One option would be to replace them with a couple of ASC PPIO cans which you can usually find on eBay. These generally are not available in values much over 100uF though, which is still probably ok for what you are trying to do.
Another option would be to invest in a pair of rubycon blackgate electrolytics which I have found to sound good in these sorts of applications - the value and voltage required would make them expensive though. (You would need a 50V cap minimum.)
Still another option if you are up for a challenge would be to switch to fixed bias operation. I don't know whether or not your transformer has a bias tap or not, but at worse you could add a small transformer to get the voltage required. Typically an EL34 (depending on brand) might require somewhere between -28V and - 36V on the grid to achieve about 60mA of idle current at around 350V - 400Vdc on the plate. Fixed bias would give you the option to tweak the operating point for your tube/transformer combo and eliminate the cathode cap altogether. Typically you would use about a 10 ohm resistor in series with the cathode connection and measure across it check and set the operating point.
In terms of A/B tests between these two amplifiers, as long as you operate both at levels well below the clipping threshold of yours you'll be ok. The most important thing of all is to make absolutely certain that the levels are closely matched. Two amplifiers of otherwise equal virtue but differing loudness will unfairly give the win to the louder of the two.
Actually I meant to bring this up, In my basement I found a printer power supply that is 30v Regulated DC, It outputs about 33vDC after all is said and done, the current rating on it is 400ma, so it should do just fine, but I am still trying to figure out what voltage to run it at? is it just the exact same voltage that i meausre across my cathode resistor?
The PSU is tiny, it is about the size of a deck of cards, and I have some extra 470u farad caps to run it through to help lower the ripple even more, and maybe run it through an LM317 to make the current exact. does this sound like a good idea?
The PSU is tiny, it is about the size of a deck of cards, and I have some extra 470u farad caps to run it through to help lower the ripple even more, and maybe run it through an LM317 to make the current exact. does this sound like a good idea?
You need a negative supply for the bias. Many ways to do this. One way is to use a common 120:12V transformer and connect it in reverse to your filament winding output. That would give you a 1:10 step-up. A bridge and some RC sections would get you a nice quiet supply. You could regulate too.
Download PSUD and play.
Sheldon
Download PSUD and play.
Sheldon
Hey, I knew that black gate capacitors were primo, however I did not know that Rubycon made them. I have two 200v 470uf Rubycon non-Black Gate capacitors I got out of a computer power supply, will these work better?
-Alex
-Alex
CFB will lower the gain a bit, which requires turning the volume up to get the same volume as before. I have not noticed any degragation in the higher frequencies from using CFB. CFB will improve the distortion and sound up until and slightly into clipping. When the amp is operated well into clipping (not always obvious on a SE amp) CFB will often increase the level of higher order harmonics even though the total distortion goes down. This can be perceived as a cloudiness of the midrange and treble. Do whatever sounds best to you. I have a switch so I can turn CFB on or off. It is usually on.
There is only one way to find out. Try them. You may also try just disconnecting the cap completely. This will lower the gain even more than CFB, but it sounds reasonably good on my amp, but I tried it in UL mode.
Kevin suggested switching the amp over to fixed bias. This will increase the power somewhat because the voltage previously dropped across the cathode resistor is now available for the tube to use. It will also improve the overload recovery. You would need a negative voltage to bias the grid. Under normal circumstances the negative grid voltage would be the same as the voltage that was across the cathode resistor. In this case the voltage across the tube will increase slightly, so you will need a little more negative grid voltage. You really want an adjustable voltage that goes from about -25 to -40 volts. This is so that you can adjust the current for each tube individually. The "automatic" feature of cathode bias is no longer present, so you must adjust the bias current each time you change tubes. The bias supply must be clean, and reliable, because if it goes away, your tubes will die an unhappy death. CFB becomes a little different with fixed bias, but it is possible.
If you have a choke already, measure its resistance with an ohmmeter. If it is lower than the resistor that you are currently using, try it. I have a large box full of chokes that I have collected over the years, but most of them have too much resistance.
I have found that the cheapest useful choke is the Triad C-24X (1 henry, 240 mA, 50 ohms) for $7.59 at alliedelec.com. The C-17X has lower resistance, and higher current capacity, but costs twice as much. I use their power transformers too. I have the 6K7VG in my amp. I get 475 volts with a SS rect, or 430 volts with a 5AR4. It is $43.
There is only one way to find out. Try them. You may also try just disconnecting the cap completely. This will lower the gain even more than CFB, but it sounds reasonably good on my amp, but I tried it in UL mode.
Kevin suggested switching the amp over to fixed bias. This will increase the power somewhat because the voltage previously dropped across the cathode resistor is now available for the tube to use. It will also improve the overload recovery. You would need a negative voltage to bias the grid. Under normal circumstances the negative grid voltage would be the same as the voltage that was across the cathode resistor. In this case the voltage across the tube will increase slightly, so you will need a little more negative grid voltage. You really want an adjustable voltage that goes from about -25 to -40 volts. This is so that you can adjust the current for each tube individually. The "automatic" feature of cathode bias is no longer present, so you must adjust the bias current each time you change tubes. The bias supply must be clean, and reliable, because if it goes away, your tubes will die an unhappy death. CFB becomes a little different with fixed bias, but it is possible.
If you have a choke already, measure its resistance with an ohmmeter. If it is lower than the resistor that you are currently using, try it. I have a large box full of chokes that I have collected over the years, but most of them have too much resistance.
I have found that the cheapest useful choke is the Triad C-24X (1 henry, 240 mA, 50 ohms) for $7.59 at alliedelec.com. The C-17X has lower resistance, and higher current capacity, but costs twice as much. I use their power transformers too. I have the 6K7VG in my amp. I get 475 volts with a SS rect, or 430 volts with a 5AR4. It is $43.
Hey, just wanted to check in
The amplifier has been running fine, I have a house full of people and everyone of them says that it sounds different than any amplifier that they have heard (solid state).
I changed to the Rubycon caps, and most definatly will not change back, as the high range sounded alot more pure, and the entire sound was better balanced.
I think in the next few days I am going to lower the Cathode resistor to 400ohms and boost the dissapation to about 23watts, between the plate and the grid. Will this be safe?
I am probably going to get some of the $17 edcor transformers its worth the cost to try UL mode, They claim 15 watts but is that true? because I know Tubelab has tested them to the limit.
right now, I have a cold!
So critical listening right now just isn't going to happen, so I have to wait a few days to change anything.
-Alex
The amplifier has been running fine, I have a house full of people and everyone of them says that it sounds different than any amplifier that they have heard (solid state).
I changed to the Rubycon caps, and most definatly will not change back, as the high range sounded alot more pure, and the entire sound was better balanced.
I think in the next few days I am going to lower the Cathode resistor to 400ohms and boost the dissapation to about 23watts, between the plate and the grid. Will this be safe?
I am probably going to get some of the $17 edcor transformers its worth the cost to try UL mode, They claim 15 watts but is that true? because I know Tubelab has tested them to the limit.
right now, I have a cold!
So critical listening right now just isn't going to happen, so I have to wait a few days to change anything.-Alex
I don't know how true it was but my old high school electronics teacher (1967) claimed that the risk of death due to electric shock was much greater if you have a cold. He believed it enough to forbid sick people from any lab work. Yes, all of our experiments used tubes.
400 ohms is too low for my amp, but you are running less voltage, so you may be OK. Check the cathode voltage, and figure the dissipation. 23 watts is OK, but remember that tube life goes down when dissipation goes up. I was running my JJ's at 27 watts with no sign of evil glow, but I don't know how long they will last at that burn rate. I am now running an old pair of 6L6GC's at 35 watts, dull glow with room lights off. The amp has been running continuously when I am home for the past week. Been running through my collection of concert DVD's. The Pulse DVD is cool. Saw that show when they played Dolphin stadium in Miami.
Sadly, the amp will have to go back into the lab tonight. I will be out of town for the next 2 to 3 weeks with limited computer access, so I won't always be able to answer.
In case you haven't figured it out yet I tend to test everything to its limits and often beyond! I just haven't had the time to do the full distortion VS frequency curves on either transformer yet, but I have ran them both in SE amplifers that made 15 watts. They are similar in sound and performance. The Edcor has the UL tap, and the Hammond has the multiple output taps. The Edcor will offer more power in UL but the ability to make a lot of power at low frequencies will still be about the same as the Hammond due to core saturation.
The power rating of any OPT is somewhat arbitrary. The Edcor and the Hammond 125CSE transformers are roughly the same size and will handle about the same power. I am sure that either one of these will pass 30 to 50 watts at 1KHz without getting hot or blowing up. The power rating of an OPT is limited by saturation at low frequencies. This is primarilly a function of the mass of the iron in its core. Neither of these will crank out much more than a few watts below 100Hz. The Edcor does sound louder, and really likes CFB, but the bass is still limited to a few (5 maybe) watts.
If you are happy with the power level of your amp but want far better bass, consider a pair of Transcendar 300B transformers. These are available on Ebay (look at item # 280013249055). I have tried these in my EL34 amp and they outperform anything near their price. In fact many people can't distinguish them from One Electron UBT3's which cost twice as much. They do not have UL taps though. If you are patient you can get these for $90 per pair. I liked them enough to buy 20 of them. I used most of them in 300B amps.
Those look like a good value, but the truth is that I am not sure if I want to run in triode mode permanently, but I have a question!
I noticed that the load on those transformers were 3k primary load
what if I take an edcore 25-16-7.6k and run it at 8 ohms, making the primary resistance 3.8k ohms. that would provide a middle ground between the current 2.5k, and 5k which I am planning to change to. would this be the best budget solution? it seems a little shadey to me personally, but I could be wrong.
on the other hand, do I even need 25w of power handling?
I noticed that the load on those transformers were 3k primary load
what if I take an edcore 25-16-7.6k and run it at 8 ohms, making the primary resistance 3.8k ohms. that would provide a middle ground between the current 2.5k, and 5k which I am planning to change to. would this be the best budget solution? it seems a little shadey to me personally, but I could be wrong.
on the other hand, do I even need 25w of power handling?
I tried running a 4 ohm load on my 5K to 8 ohm transformer to achieve a 2.5K load. They worked very good.
- Status
- Not open for further replies.