Thanks to everybody's help my Tubelab SE is running nice and reliably, but I am concerned that I'm getting about 327 volts of B+ for my 45 tubes. The tubelab manual recommends 280 or 320 volts, is 327 close enough? Now that I think about it, I measured b+ about 5 minutes after powering the unit on, I'm going to measure again after letting the unit warm up for 20 minutes or so.
I'm using a couple cheapo old 45's off ebay (one Ken Rad and one Silvertone) so I'm not going to be out hundreds of dollars if I burn them out with too much voltage, but I would like to make them last as long as I can.
The manual says I need to reduce the value of C4 to lower the voltage. Is there some formula for figuring out what value I need, or is it trial and error?
I'm using a couple cheapo old 45's off ebay (one Ken Rad and one Silvertone) so I'm not going to be out hundreds of dollars if I burn them out with too much voltage, but I would like to make them last as long as I can.
The manual says I need to reduce the value of C4 to lower the voltage. Is there some formula for figuring out what value I need, or is it trial and error?
I would say that you are OK. I get about 322V of B+ when using 45s. The vintage 45s seem like they can handle it. I'm using some older etched-base Sylvania 45s that were not the strongest tubes when I got them. I've put a few hundred hours on them maybe and they still bias about the same.
http://www.diyaudio.com/forums/tube...wer-transformer-tubelab-se-2.html#post1803717
To answer your question, the best thing is to use PSUD. Measure the HT winding AC voltage when the amp is in operation. Plug in the voltage and your capacitor and choke values and then run the simulation. Plugging a 10uF cap in the C4 position should drop your B+ about 10V or more.
http://www.diyaudio.com/forums/tube...wer-transformer-tubelab-se-2.html#post1803717
To answer your question, the best thing is to use PSUD. Measure the HT winding AC voltage when the amp is in operation. Plug in the voltage and your capacitor and choke values and then run the simulation. Plugging a 10uF cap in the C4 position should drop your B+ about 10V or more.
You will need a quality choke for such a use and you will want to add snubbers. I haven't tried it myself, but the cheap little Triad 6H choke probably won't cut it. The rule of thumb I have heard is the choke should weigh at least as much as the power transformer. Take a look at the Hammond 193 series for that. I used the 193J when I first bread boarded the Tubelab SE without C4. There was a little bit of 120Hz coming out of the speakers and the choke buzzed. It probably needed the snubbers.
I hadn't heard that rule of thumb before, not that that means anything.
The choke, in a choke input power supply, certainly needs to be at or above the critical inductance, and able to maintin that critcal inductance while supplying the desired current, but that's about it, AFAIK.
I think I have a Hammond 10H, 100ma, choke in mine, and it is stone cold silent with 45's, 46's, and 47's into 96 dB speakers.
I do get a slight amount of 120Hz hum with 2A3's and 5930's that can be heard if you get within a few inches of the speaker drivers. I believe that this is because my 5 amp regulator chip, really isn't a 5 amp part,.
I rarely use 2A3's (because I don't care for the way they sound) so I haven't swapped in a replacement part to test this theory.
Win W5JAG
BTW, what's a snubber? Is that what hi fi people call a bypass capacitor?
The choke, in a choke input power supply, certainly needs to be at or above the critical inductance, and able to maintin that critcal inductance while supplying the desired current, but that's about it, AFAIK.
I think I have a Hammond 10H, 100ma, choke in mine, and it is stone cold silent with 45's, 46's, and 47's into 96 dB speakers.
I do get a slight amount of 120Hz hum with 2A3's and 5930's that can be heard if you get within a few inches of the speaker drivers. I believe that this is because my 5 amp regulator chip, really isn't a 5 amp part,.
I rarely use 2A3's (because I don't care for the way they sound) so I haven't swapped in a replacement part to test this theory.
Win W5JAG
BTW, what's a snubber? Is that what hi fi people call a bypass capacitor?
Last edited:
I don't want to sound like I'm advocating a change from the original design.
I think there is a lot of merit to choke input supplies, but I only did it because my power transformer is a hamfest refugee, and the only way I could get the B+ where I wanted it was to go choke input.
No doubt it's better to get a correct transformer, and build the amp as Tubelab intended.
I think there is a lot of merit to choke input supplies, but I only did it because my power transformer is a hamfest refugee, and the only way I could get the B+ where I wanted it was to go choke input.
No doubt it's better to get a correct transformer, and build the amp as Tubelab intended.
Thanks for the advice. I think I'm going to leave it alone until it burns a tube up. After a year or two of running fine with junk Silvertone 45's off Ebay, I might spring for a pair of fancy new ETL or Sophia 45's. Has anybody tried these with the Tubelab?
Speaking of capacitors and upgrades, the instructions say the quality of C5 greatly impacts the sound quality of the amplifier. What would be a good example of a "high quality" cap to put there, something better than the standard part from Digikey? I've already put a motor run cap in parallel with C5.
Speaking of capacitors and upgrades, the instructions say the quality of C5 greatly impacts the sound quality of the amplifier. What would be a good example of a "high quality" cap to put there, something better than the standard part from Digikey? I've already put a motor run cap in parallel with C5.
Lowering overall ESR of the final filter improves transient response of the amp. The Panasonic cap in the parts list is a good quality electrolytic, but it can be improved with a film cap. You've done that by adding the motor run. Theoretically you can continue improving it by adding smaller film caps in parallel...the so-called 1/100 rule. Whether you will hear any difference...who knows.
To implement the 1/100 rule you would put a 1.2uf and a 0.012uf in parallel with the 120uF to achieve a more "perfect" capacitor. Personally I think the law of diminishing returns applies here. The ESR of modern caps is much lower than it was back in the day.
To implement the 1/100 rule you would put a 1.2uf and a 0.012uf in parallel with the 120uF to achieve a more "perfect" capacitor. Personally I think the law of diminishing returns applies here. The ESR of modern caps is much lower than it was back in the day.
TSE... 350v effect on 45's
I cleaned up the wiring on my TSE and put in a Mullard/Dynaco GZ34/5AR4 Copper Plate Tube I just purchased. This will be my 3rd and most expensive blown 5ar4 if it goes. With the other two tubes that gave out I was getting 328v / 338v across R30. This tube is reading 350v +/-3v. With the previous 5ar4 I tried using a 4.7uf and a 22uf in C4, attempting to lower the B+. With the 4.7uf the drop in voltage from 47uf was not that significant plus the 5ar4 (Recent music store purchase) blew. So I went with the 22uf (The 4.7 just looked too small and then I broke the can when I pulled it off the board). I am currently running the amp at 350v (across R30) with a 22uf in C4 and the 45's bias at 26 with no smoke (Yet). I have two questions: At this voltage if a tube is going to go... Which one will it be? Or, is some other component going to fail leading me to more headaches than a Newbie like me knows how to fix? My second question is that since this 5ar4 sounds really good is there another way to lower the B+ so I can use it at a safe level. I don't want to use a 5-?-4 something tube or SS to lower the voltage. In reading about rectifier tubes some say a quality tube will improve the sound. To me it seems as though my little iPod produces a larger sound stage with this specific tube and I want to keep using it... Or the whole thing could just be my imagination.
Thanks in advance for any advice.
Forgot to mention: I am using the Edcor XPWR131-120, an 80 MFD motor run, Triad C-14X choke, and the rest of the parts are stock per parts list.
I cleaned up the wiring on my TSE and put in a Mullard/Dynaco GZ34/5AR4 Copper Plate Tube I just purchased. This will be my 3rd and most expensive blown 5ar4 if it goes. With the other two tubes that gave out I was getting 328v / 338v across R30. This tube is reading 350v +/-3v. With the previous 5ar4 I tried using a 4.7uf and a 22uf in C4, attempting to lower the B+. With the 4.7uf the drop in voltage from 47uf was not that significant plus the 5ar4 (Recent music store purchase) blew. So I went with the 22uf (The 4.7 just looked too small and then I broke the can when I pulled it off the board). I am currently running the amp at 350v (across R30) with a 22uf in C4 and the 45's bias at 26 with no smoke (Yet). I have two questions: At this voltage if a tube is going to go... Which one will it be? Or, is some other component going to fail leading me to more headaches than a Newbie like me knows how to fix? My second question is that since this 5ar4 sounds really good is there another way to lower the B+ so I can use it at a safe level. I don't want to use a 5-?-4 something tube or SS to lower the voltage. In reading about rectifier tubes some say a quality tube will improve the sound. To me it seems as though my little iPod produces a larger sound stage with this specific tube and I want to keep using it... Or the whole thing could just be my imagination.
Thanks in advance for any advice.
Forgot to mention: I am using the Edcor XPWR131-120, an 80 MFD motor run, Triad C-14X choke, and the rest of the parts are stock per parts list.
Last edited:
350V B+ is pretty high for a 45.
Lowering C4 should reduce the stress on the rectifier, so I can't really explain your previous result unless there was something wrong with that cap. To lower B+, you can reduce C4 and/or increase R4 (or add a resistor in series with the choke). You can also use a larger choke with inherently more DCR, like the Hammond 158M. You can also use a different rectifier, like the 5U4.
You can add some protection to your rectifier by adding diodes in series with the plates. This is a common trick these days. The characteristics of the tube easily swamp anything that the diodes will contribute.
Lowering C4 should reduce the stress on the rectifier, so I can't really explain your previous result unless there was something wrong with that cap. To lower B+, you can reduce C4 and/or increase R4 (or add a resistor in series with the choke). You can also use a larger choke with inherently more DCR, like the Hammond 158M. You can also use a different rectifier, like the 5U4.
You can add some protection to your rectifier by adding diodes in series with the plates. This is a common trick these days. The characteristics of the tube easily swamp anything that the diodes will contribute.
Resistor in series with the choke
rknize, (Thanks for the reply)
From your suggestions, adding a resistor in series with the choke would be the easiest thing to attempt first. At the very least, it's the cheapest. What size resistor min / max would you suggest? I can rig up something to switch out various resistor values until I get an acceptable voltage (Or at least close) then maybe lower C4 again to further tweak the voltage. I guess it was possible that I got a faulty 4.7 cap???
rknize, (Thanks for the reply)
From your suggestions, adding a resistor in series with the choke would be the easiest thing to attempt first. At the very least, it's the cheapest. What size resistor min / max would you suggest? I can rig up something to switch out various resistor values until I get an acceptable voltage (Or at least close) then maybe lower C4 again to further tweak the voltage. I guess it was possible that I got a faulty 4.7 cap???
I went to R-Shack today and all they had were 50ohm 10w resistors. (The place isn't what it use to be). One pkg got me 2pcs. Before I started I turned on the amp and was reading 362v across R30. Check the wall outlet and it read 125.7v about 6 volts higher than normal (Don't know if it makes a difference or not). Put both resistors in series with the choke and the voltage dropped to 341v +/-1. Not as much as I was hopping for.
Is my thinking correct in that changing to the Hammond 158M is going to yeild me approx. a 30v or so drop? I don't understand the math? It's not my wishful 300v but, 330v is better than 350v. Will the higher DCR degrade the sound (I thought I read that on a post???)
I'll also try another 4.7uf in C4 to give me a few less volts? Which leads me to... "How low can I safely go in C4 before there is no reason to have one? I believe someone posted that he went down to 0.1uf and another removed C4 totally to achive the voltage they wanted.
I really appreciate your help.
Thank you
***The amp has been running for an hour or so and it appears to have settled at 337v
Is my thinking correct in that changing to the Hammond 158M is going to yeild me approx. a 30v or so drop? I don't understand the math? It's not my wishful 300v but, 330v is better than 350v. Will the higher DCR degrade the sound (I thought I read that on a post???)
I'll also try another 4.7uf in C4 to give me a few less volts? Which leads me to... "How low can I safely go in C4 before there is no reason to have one? I believe someone posted that he went down to 0.1uf and another removed C4 totally to achive the voltage they wanted.
I really appreciate your help.
Thank you
***The amp has been running for an hour or so and it appears to have settled at 337v
Yes, you can completely remove C4. The CLC pi filter then becomes an LC filter. I did exactly this when I first tested the TSE using a power transformer that I had on hand.
Switching to an LC filter puts a lot more strain on the choke, as it has now become the first stage of the filter. The little C-14X is probably not up to the task. Even the Hammond 193J in that picture struggled (buzzed). It was effective, though. From the perspective of the rectifier, an LC choke is ideal because the rectifier won't see large current spikes on each cycle.
Switching to an LC filter puts a lot more strain on the choke, as it has now become the first stage of the filter. The little C-14X is probably not up to the task. Even the Hammond 193J in that picture struggled (buzzed). It was effective, though. From the perspective of the rectifier, an LC choke is ideal because the rectifier won't see large current spikes on each cycle.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.