Tubelab SE amp question

[smbrown]

Finally got around to powering up the board I completed a couple months ago. Too many projects. I'm using it with 300Bs (Electro Harmonix) and I've got a problem. I think I know the source of the problem, but appreciate guidance on a fix. It is that after things warm up I lose the heater voltage on the 300Bs and I suspect the difference at the front of the regulator (close to 6.6v) and the 300B (about 4.8v) is far enough apart that it's causing the regulator to heat up and shut down. I put some series resistance in the filiment supply line so that my input is now 6.3 or close, and it still shuts down, but takes much longer. However, if I put on an underrated transformer, say a 2a 6.3v tfr, it runs forever, no problme except that the filiment transformer gets extremely hot. I think if I can raise the output of the regulator to close to 5v the drop would be smaller and not cause an issue. By the way, the board is currently not enclosed in anything, and the room is normal temp, so it isn't that heat is building up. What I need some advice on is what components to change to what values on the reg to help solve this. I suspect the ratio of r2 and r35 is off, but not sure which direction to go. Thanks!!

[tubelab.com]

4.8 Volts is a bit lower than I usually see on these regulators. They usually run 4.9 to 5.1, but it is probably OK. The Sharp regulator does have an internal temperature shutdown circuit that will turn the chip off if its internal temperature gets too high.

The chip will operate fine with up to 7 volts on its input, but some additional heat sinking is required beyond the board mounted heat sink. The board mounted heat sink is adequate for 45 tubes with good airflow. A larger heat sink is required for 300B's or 2A3's. I usually use the heat sink removed from a CPU chip out of an old computer.

Reducing the voltage on the input of the IC may cause it to operate in or near the dropout region. This will reduce its ripple rejection and may cause hum.

[smbrown]

George, thanks, and that's what I suspected. Not anxious to strap in a big heatsink, however... nor do I have one handy. And yes, if I lower the input too far it does result in hum. Sounds as if I'm not in a happy place. Are others who use 300B's having to resort to these meaures?

[tubelab.com]

Are others who use 300B's having to resort to these meaures?

The need for suplemental heat sinking has been mentioned several times on the forum and it is mentioned on my web site, but probably not as prominently as it should be. I am in the process of updating the instructions to include new information. I am also finishing a new 300B based Tubelab SE build up that will be added.

This has gotten worse since I designed the amplifier because the transformers we get today (especially Hammonds) tend to put out more voltage than their specs show, and our line voltage keeps increasing. Mine is about 123 volts.

You will see a big green one in my Lexan amp. It was robbed from an old Pentium 1 computer and is overkill for most situations, but required in that small closed box. The board shown in the setup and testing page uses a heat sink from a Pentium 3, I removed its fan. It is about the right size. If you have good ventillation a small piece of flat or angle aluminum may be sufficient. I have also used a small piece of angle aluminum that was screwed to the chassis with good results.

[smbrown]

Thanks for the note. I've added a small Radio Shack heat sink to the back of the standard one and some beefier hardware in the hopes of getting rid of the additional heat. I also switched the filiment transformer to a 3A unit. Now I get 6.32vac and after changing (I think) R35 from 255 to 249 ohms, I get 5.01 v at the 300Bs. I've not had a chance to run it for more than a few minutes but I will do so tomorrow. In the mean time, let me just say that this amp has the best bass of ANY SE amp I've ever heard. The soundstage is great, image placement is rock solid and overall it sounds very nice. I still need to give it a dozen hours of break in, but if what I'm hearing after a couple hours is a good indication, this will be a great sounding amp.
By the way, the PT I'm using is a replacement one for a Dynaco ST35. It is rated at 660vct and 180ma. There are 2 6.3v 2.5 amp filiment windings. I'm using one for the 5AR4 (with a dropping R) and curently NOT using the other one due to the heat build up in the transformer. Hence the need for a seperate filimen TFR. I have a 700vct transformer I'd like to try, do you think that would make the B+ too high? I'm guessing close to 440. I was thinkin I could sub in a 5U4 and be okay due to the larger V drop across that rectifier ( the other tfr has a 5v winding that would more than accomodate a 5U4). I betting I'd end up in the 420v range.

[tubelab.com]

do you think that would make the B+ too high? I'm guessing close to 440.

The B+ would likely be too high to apply to a 300B tube. The higher secondary voltage would raise the negative supply voltage too much. This creates a heat issue for the droping resistors and the mosfets.

I use a Hammond 272JX in my latest 300B amp. It is rated for 600VCT but like most Hammond transformers I get more than that. I get 660VCT giving me a B+ of about 365 to 370 volts depending on the line voltage. That is just about right for most 300B tubes. I have found some 300B tubes (older Sovteks) that will go into runaway if operated at 400 volts even though they claim to be rated for 450 volts.

[BillEpstein]

Hi Steve and George.

Verrry timely thread for me! I couldn't find a 376X anywhere and the Edcors, I dunno...power transformers should be black! I just ordered a JX for my SE. That and a pair of Magnequest RH-80s: I'm good to go for magnetics.

[smbrown]

Hey Bill, great that you're building stuff! I really like this little amp. I've got a pair of ELM 300B's in there right now (tight fit!), and I'm using my favorite James OPTs. Been thinking about taking the plate choke and transformer out of the 45 amp and seeing those those sound in the circuit. I think the plate choke is good for 60ma, IIRC. Let me know how your project sounds - oh, and if you put it on a breadboard, watch the cat around all that exposed HV!

[BillEpstein]

I still have the pair of EXO-045s that we used as chokes originally (BCP-15s in there now) so I could give that a shot, too.

At first I thought I'd go for the 45s to see how it compared to the Simple but now I'm going straight to 300Bs so I'll have the Simple SE 6CA7, SE 300B and your 45. A Trifecta