• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Joining the Tubelab SE club

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Ok, plugged in the brand new Sovtek 5AR4, re-biased the output tubes, and let things settle a little. And now my B+ is even lower! More like 330-335V.

So, it seems like I should next try swapping out C4 for a higher capacitance (I have one that's 56uF instead of 47 as is in there now). This is still below the max of 60uF in the 5AR4 data sheet; is there some other reason I shouldn't go to a higher capacitance here? Is there empirical evidence that even >50uF will strain the 5AR4?

Or is ~345V of B+ sufficient and I should leave it as-is? I'm trying to make a decision here but there doesn't seem to be a straightforward answer (at least not that anyone's willing to put in writing ;)). George mentions in numerous places that the B+ should be higher, so it seems to me that's more important.
 
So, it seems like I should next try swapping out C4 for a higher capacitance (I have one that's 56uF instead of 47 as is in there now). This is still below the max of 60uF in the 5AR4 data sheet; is there some other reason I shouldn't go to a higher capacitance here? Is there empirical evidence that even >50uF will strain the 5AR4?
56<60 so what's to worry?
 
It's getting close to the limit and current-production tubes are well-known not to handle being pushed to the old specs. An NOS tube will probably be fine with more C4, but I wouldn't do it with a current tube. My Dynaco has 50uF there and it eats JJ's for breakfast until I put some UF1000s on there to help out.

The SSE doesn't draw as much current overall, so it might fair better. However from what I recall looking at PSUD, the increase in B+ from 47uF to the max made very little difference at these currents. Sorry the new tube didn't fix it. Maybe the lower DCR of the choke is what is making the difference for me. According to Mr. Ohm, if we assume I was drawing 60+60+12+12=144mA and my choke's DCR is really 82 ohms, that means I was dropping about 12V across it. For your 150 ohm DCR choke, that would be about 22V. So that accounts for 10V, at least.
 
Ok, thanks. Well, I'll try putting in the 56 cap and see what difference it makes. I can also change the choke easily enough... are there alternatives to the one you're using? It's pretty expensive. Plus yours is 10H, mine is 6H... does this matter? Why'd you go with 10H?

Edit: playing in PSUD, it seems 6H gives much better damping, with 10H it initially overshoots B+ (by ~40V) before going back down. No idea how realistic this is, though, with warmup times for the rest of the system's components. But the inductance doesn't change the final B+ much at all, but (as you say) the DCR does.
 
Last edited:
Plus yours is 10H, mine is 6H... does this matter? Why'd you go with 10H?

I had a couple of them from an abandoned project, so I used them. In the pic above, you can see it front and center laying on its back. I took the bells off and put them on the Edcor XPWR131. They fit perfectly and reduced the overall height slightly, which I needed. The Hammond bells also have the round 1/2 EMT knock-outs, which I used in this case.

Edit: playing in PSUD, it seems 6H gives much better damping, with 10H it initially overshoots B+ (by ~40V) before going back down. No idea how realistic this is, though, with warmup times for the rest of the system's components. But the inductance doesn't change the final B+ much at all, but (as you say) the DCR does.

The overshoot is nothing to worry about...it doesn't happen in practice because of the soft-start of the rectifier. However you can interpret it as a sign that the supply is not critically damped. Simulate any vintage design and you will see this (usually much worse). In an SE amp in class A, this doesn't matter because the amp is drawing about the same current all the time. If you drive it hard so that it goes into cut-off a lot, then it becomes more of a concern. However at that point an SE amp is distorting like crazy, so it's still sort of moot.

I happen use a 220uF in the C5 position (had it on hand) and it swamps-out most of the overshoot anyway.

A cool way to test how damped your supply is in PSUD is to use the two-stage constant current load. Set it to your predicted idle load 2 seconds and then full load for 2 seconds (or vice-versa). Look for ringing when the transition happens. This is more important with PP amps if it is expected to operate in class AB1 sometimes. It's even more important in AB2, obviously.
 
So, I swapped out C4 with the 56uF cap, and guess how much difference it made? Zero. Exactly the same B+ as I was getting before, 345V (back to the the GE rectifier). I'm confused as to why this makes no difference at all. Maybe because of the big motor run cap (100uF) in parallel with C5? Though I wouldn't expect that to affect B+ much... I may try taking it out to see what happens, just for kicks.

It's frustrating to not know why the values I'm seeing aren't the same as other folks, why it doesn't seem to be behaving the same way. Maybe everything's just fine, but I'm still confused. (I even tried a different voltmeter to make sure it wasn't that acting up.) I just want to be sure this doesn't indicate some other sort of problem somewhere.
 
Ok, maybe this is a stupid question, but this B+ measurement is supposed to be taken with the system up and running with full bias current, right? Because the B+ I see is highly dependent on where I set the bias current; I've been keeping it around 60mA for now (new tubes), but if I push it up closer to 70 for both tubes, B+ drops by at least 5V. And if I go the other way, turning the bias pots back down so that there's zero bias current, then B+ goes up to ~420V. Is this to be expected?

I do notice that my B- (bias supply, at R7) is around -295V, this seems extreme, given that the schematic says -150V... Is this also to be expected?

Edit: I tried it without the motor run cap, no change in B+, not surprising there.
 
but this B+ measurement is supposed to be taken with the system up and running with full bias current, right?.......so that there's zero bias current, then B+ goes up to ~420V. Is this to be expected?

Yes, it is normal for any unregulated power supply to drop under load. Tube rectified supplies drop more than SS rectified supplies.

I do notice that my B- (bias supply, at R7) is around -295V, this seems extreme, given that the schematic says -150V... Is this also to be expected?

Yes, again. The -150 volt readings were taken on a 45 based amp with a smaller power transformer.

the 56uF cap, and guess how much difference it made? Zero.

I have not had the time to post here recently but I need to explain that the value of C4 will affect the B+ voltage up to a point. Small values (2 to 20 uF) can be used to reduce the B+ voltage if your transformer creates too much B+. Once you reach the limit larger values will only make your rectifier tube unhappy. I put a 100uF in my Lexan amp in an attempt to squeeze a few more volts out of a small power transformer. It did not make any difference over a 47uF. That amp runs the 300B's at about 295 volts and it sounds very good, but only makes about 4.5 watts per channel. I usually run 45's in that amp where I get about 320 volts of B+ due to the much lower bias current (30mA).

I have another amp using 300B's and a Hammond 272JX. The B+ voltage used to be in the 380 to 390 volt range but now I get about 345 to 355 volts. What changed? The biggest change happened 2 years ago when the power company replaced the 30 year old large metal cased pole transformer (lightning zapped it) with a new one that is plastic cased and much smaller. Now the Hammond emits a faint buzz where it didn't before, and the B+ voltage is lower. The B+ voltage took another dive over the past 6 months as everyone got new flat screen TV sets. The 60 Hz sine wave coming from my wall outlet now has visible flat topping and measures between 8 and 12 percent distortion depending on the time of day.

Or is ~345V of B+ sufficient and I should leave it as-is? I'm trying to make a decision here but there doesn't seem to be a straightforward answer (at least not that anyone's willing to put in writing ). George mentions in numerous places that the B+ should be higher, so it seems to me that's more important.

If you are happy with the sound of your amp, then by all means leave it alone and enjoy it. One of the big "secrets" about the 300B tube is that it has a rather large sweep spot (it sounds good over a large range of operating conditions). I have not noticed any change in the sound of my amp as the B+ dropped over the past 2 years. The output power probably went down 1/2 to 1 watt, but only the measuring equipment will find it. I only noticed the change because I had the unfinished amp on the bench when the pole transformer blew and I noticed the loss in B+ the next day. My Simple SE lost a few volts too. The amp is still unfinished and I have taken it appart several times to work on the cabinet only to put back on the plywood breadboard for listening. It is currently apart again. I really will finish it soon, but I have some Simple P-P's to build.
 
Thanks George, I appreciate your time and detailed response. So, it sounds like I have nothing to worry about. I'll probably put the 47 C4 back in and just go with it. Anyone have a feeling for whether it's better to use a higher bias current but lower B+, vs the other way around?

I don't know how the amp sounds yet... still hooked up to crappy test speakers. I've been afraid to put in the good speakers because I wasn't sure the amp was behaving right. (And how do I know it's behaving right? George says, because it sounds good. ;) Chicken or egg...) But yes ultimately I understand the final measure of the amp is the sound, it's just that the scientist in me wants an objective quality assurance measurement first. ;)

How would one increase B+ then in a more general, put in a higher-voltage power supply? Or are there limits to the voltages that the rectifier and other existing components can handle, such that a larger PS would require more substantial redesign? Just wondering... not planning to do this myself!
 
The near-term practical limit is going to be the voltage rating of the caps in the B+ and B- supplies. Then you need to start adjusting the dropping resistors in the bias circuits for both B- and the FETs. Next up would be the rectifier, but that is well out of the range of the output tubes anyway.

If you want to eek out another 10 volts or so, you can try a different choke. Another trick would be to unload the 6.3V winding on the Edcor and use a separate transformer. That would free-up some capacity on the primary and the core which should give the other secondaries a little boost. Of course, that would sort of negate the point of that transformer. :) You could also go to a silicon rectifier.

BTW, how does your 5V look?

Like George implied, it's probably not going to make a noticeable difference to your ears with the 300Bs.
 
Ok, looking around for a possible diferent choke, 6-10H range, 200mA, <100ohm DCR... anybody know of a relatively inexpensive source?

What's the actual current coming out of the B+ supply? Is a 150mA choke sufficient? I guess I could measure this for myself... may do that tonight as I start tinkering around, hopefully starting work on the enclosure that I've been too lazy to get to. Sigh...
 
Hello guys,
My name is Pavel, I`m from Columbus Ohio.

I have some experience in DIY, I built speakers and amp, not tubeamp, just stereo double mono (spent more than 400$ its not Chinese kit)). But I never heard how sound tubeAmp in the real, only a lot of good reviews and words. And I wanna try build SSE, but before spend another hundredths dollars will be cool hear amp in real.
Maybe we have member in Columbus area and we will able compare my amp and SSE?
Thanks.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.