Hi,
gorgeous project!
Some thoughts on it, and answers to some questions:
The SVT's final stage is not an Ultralinear design, just straight pentodes. You could modify your amp to get it switchalbe between both modes, and even a third one, the triode mode, by using a dual pole triple throw switch of sufficient voltage capability (!). Just connect all three screens of every push-pull side to both the wipers and the three fixed contacts to Eb (for pentode), or UL taps (for Ultralinear) or plates (for triode mode), respectively. Also the SVT doesn't use an interstage tranny, as far as I know.
I'd never replace 6L6GC's by 6550's ot KT88's, due to their bigger heater current (1.6 vs. 0.9 amps) requirements which most likely will burn the mains transformer's heater winding.
I'd ever replace any carbon resistor by it's modern, and reliable metal film equivalent. There might occur some sound change, due to the fact that you bring it back to it's factory sound when the original carbon rsistors still had their original values.
I'd also ever replace all electrolytics. Due to ageing their capacitance values are prone to decrease, while leakage current most likely will increase to unwanted amounts.
I'd insert 10 ohms resistors into the cathode lead of every final tube. Thus you can "match" them by your own to a sufficient level for (bass) guitar amplification.
Last, but not least: As you intend to use it as an bass guitar amp, I'd replace the five capacitors in series with the inductances in the equalization circuitry by some of four times the original value, thus bringing the equalization ranges an octave down.
Best regards and good luck!
Addendum: In real pentode mode your amp will deliver even more output power, but distortion will increase somewhat.
gorgeous project!
Some thoughts on it, and answers to some questions:
The SVT's final stage is not an Ultralinear design, just straight pentodes. You could modify your amp to get it switchalbe between both modes, and even a third one, the triode mode, by using a dual pole triple throw switch of sufficient voltage capability (!). Just connect all three screens of every push-pull side to both the wipers and the three fixed contacts to Eb (for pentode), or UL taps (for Ultralinear) or plates (for triode mode), respectively. Also the SVT doesn't use an interstage tranny, as far as I know.
I'd never replace 6L6GC's by 6550's ot KT88's, due to their bigger heater current (1.6 vs. 0.9 amps) requirements which most likely will burn the mains transformer's heater winding.
I'd ever replace any carbon resistor by it's modern, and reliable metal film equivalent. There might occur some sound change, due to the fact that you bring it back to it's factory sound when the original carbon rsistors still had their original values.
I'd also ever replace all electrolytics. Due to ageing their capacitance values are prone to decrease, while leakage current most likely will increase to unwanted amounts.
I'd insert 10 ohms resistors into the cathode lead of every final tube. Thus you can "match" them by your own to a sufficient level for (bass) guitar amplification.
Last, but not least: As you intend to use it as an bass guitar amp, I'd replace the five capacitors in series with the inductances in the equalization circuitry by some of four times the original value, thus bringing the equalization ranges an octave down.
Best regards and good luck!
Addendum: In real pentode mode your amp will deliver even more output power, but distortion will increase somewhat.
Last edited:
Kay, I can't make out exactly what you are saying. If I changed tube type to one that required more heater current its not a big deal to make a seperate heater supply just for the outputs. This amp is new enough to have modern caps. The main caps have been replaced. I do intend to do some work on the power supply filter ladder though, until an unfiltered scr light dimmer injecting line noise doesn't affect the amp (thru the 115 VAC; the RF noise from a dimmer radiated thru the air from the dimmed light lines is a seperate subject).
I may try pentode mode! But for now I have a lot of other things to address first.
I may try pentode mode! But for now I have a lot of other things to address first.
Hi,
yes, if you'd consider replacing yout output tubes by 6550's, you'd need an additional heater supply. Also don't forget that both tubes require different plate-to-plate load impedances.
As they were mentioned above: EL34's are not a good substitution! The also require more heater current (1.5 A vs. 0.9 A) than 6L6GC's and screen voltage in this UL circuitry most likely will exceed the maximum rating given for these tubes. Unless you can buy a sextett of NOS tubes (from Mullard, Telefunken etc.) for a huge amount of money, as these really are capable of the higher screen voltage, you'd better forget that idea.
You could double both the inductors' and the capacitors' values within the EQ circuitry, Q then should remain the same. But I guess that you don't even know the inductors' values, so changing the caps should be the best, and maybe the only possible, way to start with. Finally, listening experience tells you whether you're on the right way.
Best regards!
yes, if you'd consider replacing yout output tubes by 6550's, you'd need an additional heater supply. Also don't forget that both tubes require different plate-to-plate load impedances.
As they were mentioned above: EL34's are not a good substitution! The also require more heater current (1.5 A vs. 0.9 A) than 6L6GC's and screen voltage in this UL circuitry most likely will exceed the maximum rating given for these tubes. Unless you can buy a sextett of NOS tubes (from Mullard, Telefunken etc.) for a huge amount of money, as these really are capable of the higher screen voltage, you'd better forget that idea.
You could double both the inductors' and the capacitors' values within the EQ circuitry, Q then should remain the same. But I guess that you don't even know the inductors' values, so changing the caps should be the best, and maybe the only possible, way to start with. Finally, listening experience tells you whether you're on the right way.
Best regards!
I bought another 6L6. It's a JJ and not identical to either type I have now. I'm planning to convert to bias pots ASAP. Anyway, it doesn't get as hot now and works great. So I'm sticking to 6L6GC for now, while I address more pressing issues. I really want to move that 'distortion' stage to become another switchable gain stage w/ volume, located between the tone circuits and the EQ circuits.
Then I've got other heads to reassemble! And I should have at least one real bass bottom with with some larger low-frequency speaker in a bigger box, as alternative to all these open-back 4x10 and closed-back 4x12.
Then I've got other heads to reassemble! And I should have at least one real bass bottom with with some larger low-frequency speaker in a bigger box, as alternative to all these open-back 4x10 and closed-back 4x12.
teee Heee... rebuilt My '64 Fender Bassman, a few years back. Replaced ALL carbon comps with 1% carbon films, fitted all fresh caps Sprauge Atoms etc. etc.
Oem resistors had seriously drifted. Oem caps were either leaking already or close to it.
Utah speakers were "fine' as were the OEM Sylvania's
Sound.. was... different as result of it being back to factory circuit spec...clearer, brighter with far less brownout, notably louder too.
Took about a week to get used to the as new sounds. However all who have since listened / used it prefer the restored 'as Leo intended' voicing.
The Weenies can moan about it's not sounding Clapped out vintage.
Ohhh but wait.. It doesn't belong to them
Oem resistors had seriously drifted. Oem caps were either leaking already or close to it.
Utah speakers were "fine' as were the OEM Sylvania's
Sound.. was... different as result of it being back to factory circuit spec...clearer, brighter with far less brownout, notably louder too.
Took about a week to get used to the as new sounds. However all who have since listened / used it prefer the restored 'as Leo intended' voicing.
The Weenies can moan about it's not sounding Clapped out vintage.
Ohhh but wait.. It doesn't belong to them
Last edited:
Well, this sure isn't a bluesy saggging-power distorting amp wheezing out an emotive cry. It's not a primary guitar amp, it's a clean machine primarily for bass and occasional guitar duty, though it might have more tone potential later if I add switches for non-ultralinear mode, or to remove some feedback. And of course if I move that extra pre-amp stage to between the "tone" section and the "EQ".
So I picked up a huge inductor in an attempt to get rid of every bit of power amp hum, yet not drop B+ much in a retrofit application. It's truly huge and really heavy, so I'm investigating many mounting alternatives. I should run over there and measure the DC resistance, as I bought it without knowing, just figuring based on its size that it must be low. And with the big inductor and more caps I will need to upgrade the rectifier or it will blow at turn-on. Turn-on is going to stress the transformer more than it's ever been pushed before, so I'm not going overboard with the size of the caps.
Now...it models with steady load, but should I be thinking about dong anything to dampen any resonances in the power supply due to the capacitance and inductance I'm adding?
So I picked up a huge inductor in an attempt to get rid of every bit of power amp hum, yet not drop B+ much in a retrofit application. It's truly huge and really heavy, so I'm investigating many mounting alternatives. I should run over there and measure the DC resistance, as I bought it without knowing, just figuring based on its size that it must be low. And with the big inductor and more caps I will need to upgrade the rectifier or it will blow at turn-on. Turn-on is going to stress the transformer more than it's ever been pushed before, so I'm not going overboard with the size of the caps.
Now...it models with steady load, but should I be thinking about dong anything to dampen any resonances in the power supply due to the capacitance and inductance I'm adding?
I measured the inductor. It's less than 7 ohms. 1.35 Henries. 1.1 amps. Teseted voltage: 1780. From Submarine Signal Co., Boston Mass. Probably meant to be used as ballast on a submarine; it's probably 20 pounds LOL.
That 7 ohms figure means a really minimal drop in B+ voltage from adding this choke. Simulation says RMS mighht only drops about 4 volts. That's incredible, and gets rid of almost all the ripple.
But the head is starting to weigh like an SVT.
That 7 ohms figure means a really minimal drop in B+ voltage from adding this choke. Simulation says RMS mighht only drops about 4 volts. That's incredible, and gets rid of almost all the ripple.
But the head is starting to weigh like an SVT.
Hi,
*imho* there's no need to filter all the current by such a bulky choke. Just filtering the screen and preamp supply current will do, due to the plate common mode rejection of the push-pull output stage.
Increasing the filter cap values is a good idea in an amp for the bass guitar. Just replace them by 470 µF each. The original rectifier diodes supposedly will handle the inrush current, as even standard 1N4007's are rated for a peak surge current of 30 A.
Best regards!
*imho* there's no need to filter all the current by such a bulky choke. Just filtering the screen and preamp supply current will do, due to the plate common mode rejection of the push-pull output stage.
Increasing the filter cap values is a good idea in an amp for the bass guitar. Just replace them by 470 µF each. The original rectifier diodes supposedly will handle the inrush current, as even standard 1N4007's are rated for a peak surge current of 30 A.
Best regards!
No doubt you're right; this is indeed very eccentric. It's the American in me: nothing exceeds like excess. If I hadn't made the cabinet extra-large it might not even fit. It certainly wouldn't be very portable for one person anymore, but it was never well balanced because one transformer was heavier than the other; I hope I can offset that weight imbalance and not make it any worse. The cabinet should really have handles at each end. Good thing I added aluminum base plates under the handle end mounts. At least I didn't pay new list price for the choke; the darn thing appears to have heavy cast ends too. It was probably intended for a base transmitter of some sort I would suspect. There's not a lot of selection available in these sizes of chokes, or I could have gotten away with a third less weight easily, though there are even larger ones on eBay. It will be a very interesting experiment; the weight is indeed the penalty. I will probably mount it so that it's easily removable with a plug and replaced with a small-value high-wattage wirewound resistor. I certainly could have gone farther to balance the push and pull sides to minimize the hum and then added a choke for the splitter.
There will be additional chokes further down the filter ladder for the lower-voltage taps.
Now I'm pricing new larger elctrolytic caps at Mouser and I'm shocked at the prices! There are cheaper caps, but not in stock. Perhaps someone can make some recommendations. Many have no ESR spec. Many are rated for 550V, which may be calling it close in a high-temperature 500v application.
And I'd also like to upgrade the rectifier to a modern 1-piece bridge unit that will handle the start-up current, mounted on a small heatsink of its own, mounted lower in cooler air and away from the tubes, instead of adding any heat to the chassis. And I'll probably put some ceramic disc caps right across the bridge, like the stock diodes had.
I don't really want to kill the power transformer either, trying to fill huge electrolytics and chokes at start-up.
I assume the capacitor values (220 stock main supply I think) in the schematic which are unlabelled are microfarad, not millifarad. I always find the mfd versus ufd for microfarad and Mfd for microfarad oddly inconsistent and confusing.
I think I'll also add some internal rail fuses.
There will be additional chokes further down the filter ladder for the lower-voltage taps.
Now I'm pricing new larger elctrolytic caps at Mouser and I'm shocked at the prices! There are cheaper caps, but not in stock. Perhaps someone can make some recommendations. Many have no ESR spec. Many are rated for 550V, which may be calling it close in a high-temperature 500v application.
And I'd also like to upgrade the rectifier to a modern 1-piece bridge unit that will handle the start-up current, mounted on a small heatsink of its own, mounted lower in cooler air and away from the tubes, instead of adding any heat to the chassis. And I'll probably put some ceramic disc caps right across the bridge, like the stock diodes had.
I don't really want to kill the power transformer either, trying to fill huge electrolytics and chokes at start-up.
I assume the capacitor values (220 stock main supply I think) in the schematic which are unlabelled are microfarad, not millifarad. I always find the mfd versus ufd for microfarad and Mfd for microfarad oddly inconsistent and confusing.
I think I'll also add some internal rail fuses.
Yeah, I'm probably crazy for trying this on a 180-watt amp instead of a 50-watter first.
I don't think the stock diodes are in the selection drop-down list in the simulator. Then again, I doubt this simple simulator really simulates the effect of the duration of the current surge on the diodes realistically, it just tells you when the single 'spec' is exceeded.
I was modelling a 330 uF cap before the choke and another 330uf cap after the choke. That's using one 550v electrolytic at each point instead of two 260v caps in series like the stock config. Then, desipte the increased energy, I was hoping to use drain resistors with higher resistance, since they didn't also have to divide the voltage evenly across two caps in series anymore. I was hoping that might make up for a bit of the voltage lost across the new additional filter.
I don't think the stock diodes are in the selection drop-down list in the simulator. Then again, I doubt this simple simulator really simulates the effect of the duration of the current surge on the diodes realistically, it just tells you when the single 'spec' is exceeded.
I was modelling a 330 uF cap before the choke and another 330uf cap after the choke. That's using one 550v electrolytic at each point instead of two 260v caps in series like the stock config. Then, desipte the increased energy, I was hoping to use drain resistors with higher resistance, since they didn't also have to divide the voltage evenly across two caps in series anymore. I was hoping that might make up for a bit of the voltage lost across the new additional filter.
At least I didn't try to find a 10 Henry inductor to add before the main rail line to the output trans center tap!
My first inclination was a much smaller 1/10 Henry between larger caps, but that didn't help much according to the simulator. Then again, this simulator doesn't show real-life garbage on the line, and the resolution of the view of the ripple just keeps getting better and better, but the ripple never blends into the background noise because there isn't any simulated signal or noise injected onto the line, just the 120 Hz from the rectifier.
I'm curious how much ripple from the stock supply gets cancelled in the stock output stage. Seems to me that's also voltage and current wasted, power ending up as heat in the output tubes just to make more heat in the output transformer. Perhaps reducing the cancellation occurring in the output transformer a bit might help it survive the additional power from the power supply upgrade. Then again, that's probably few watts and at these levels it takes many to make an audible increase in db volume.
My first inclination was a much smaller 1/10 Henry between larger caps, but that didn't help much according to the simulator. Then again, this simulator doesn't show real-life garbage on the line, and the resolution of the view of the ripple just keeps getting better and better, but the ripple never blends into the background noise because there isn't any simulated signal or noise injected onto the line, just the 120 Hz from the rectifier.
I'm curious how much ripple from the stock supply gets cancelled in the stock output stage. Seems to me that's also voltage and current wasted, power ending up as heat in the output tubes just to make more heat in the output transformer. Perhaps reducing the cancellation occurring in the output transformer a bit might help it survive the additional power from the power supply upgrade. Then again, that's probably few watts and at these levels it takes many to make an audible increase in db volume.
The capacitor multiplier idea somebody proposed for the second filter was very interesting, but I'm probably not quite ready for that yet.
I'd imagine using a capacitance multiplier circuit must affect the selection of the capacitor...wouldn't it have to drain & fill a lot more? Seems to me that's the heart of the matter, in a standard filter just a little ripple current and class B output current (and replenished from the trans/rect) flows in & out of the cap, but in a capacitance multiplier circuit a lot more current flows in & out of that cap when it's simlating a larger value.
I'd imagine using a capacitance multiplier circuit must affect the selection of the capacitor...wouldn't it have to drain & fill a lot more? Seems to me that's the heart of the matter, in a standard filter just a little ripple current and class B output current (and replenished from the trans/rect) flows in & out of the cap, but in a capacitance multiplier circuit a lot more current flows in & out of that cap when it's simlating a larger value.
Cyclecamper,
do I understand you well that you fear to dissipate additional heat by cancelling ripple in the output stage? No way - this is done automatically by the tranny if the sums of all plate currents on each p-p side are equal! No heat ever produced by this! But you indeed will produce additional heat if you try to reduce ripple in the supply current flowing into the output tranny's center tap, more if you use an ohmic resistor (C-R-C filter), less for a choke (C-L-C), both for a given hum reduction rate. *imho* hum reduction will be sufficient, for not to say excellent, even in the original supply circuitry, if you provide a separate grid bias pot for each output tube, or at least for each p-p side.
Which are the prices of the electrolytoics you're looking for? Here in Germany those aren't that expensive, I think. If you do understand a little German, have a look here, e.g. We can buy electrolytics even cheaper from surplus here.
Another thing to mention: Induced by this thread I've been looking for such an amp on eBay. Indeed, a German guy offered one these days, calling € 850 (~$ 1120) - way too much. Of course he didn't sell it. At the end of the auction I offered him € 500 (~ $ 670). I'm still looking forward to his answer ;-).
Best regards!
do I understand you well that you fear to dissipate additional heat by cancelling ripple in the output stage? No way - this is done automatically by the tranny if the sums of all plate currents on each p-p side are equal! No heat ever produced by this! But you indeed will produce additional heat if you try to reduce ripple in the supply current flowing into the output tranny's center tap, more if you use an ohmic resistor (C-R-C filter), less for a choke (C-L-C), both for a given hum reduction rate. *imho* hum reduction will be sufficient, for not to say excellent, even in the original supply circuitry, if you provide a separate grid bias pot for each output tube, or at least for each p-p side.
Which are the prices of the electrolytoics you're looking for? Here in Germany those aren't that expensive, I think. If you do understand a little German, have a look here, e.g. We can buy electrolytics even cheaper from surplus here.
Another thing to mention
: Induced by this thread I've been looking for such an amp on eBay. Indeed, a German guy offered one these days, calling € 850 (~$ 1120) - way too much. Of course he didn't sell it. At the end of the auction I offered him € 500 (~ $ 670). I'm still looking forward to his answer ;-).Best regards!
Hmm, let me think. OUch. I'll have to think more later. That just doesn't sound right. I would think B+ ripple current is turned into opposed currents at the two end taps of the output transformer primary, which the transformer primary winding resistance turns into heat. Of course it's not just resistance...lots of inductance there...and of course no output.
I see serviceable new high-voltage electrolytics for reasonable prices, but never in stock! Then ones that are in-stock for high prices. And none with complete specs.
I know of two Super Twins here in the US, one for $399 and another for $499. Definitely a bargain.
I see serviceable new high-voltage electrolytics for reasonable prices, but never in stock! Then ones that are in-stock for high prices. And none with complete specs.
I know of two Super Twins here in the US, one for $399 and another for $499. Definitely a bargain.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.