Tube Amp Power Supply Design
Vložit
- čas přidán 7. 07. 2024
- Link to the example circuits -- drive.google.com/drive/folder...
Link to our FB page ----- / bugamps
Link to our website ----- www.bugamps.com
Warning!
The circuits discussed in this
video contain High Voltage.
There is a risk of injury and death
when working with these circuits.
Please take precaution.
Don't attempt to build these circuits if you
are unsure of your ability to do so safely. - Věda a technologie
The topics that you cover throughout your lecture are what I would’ve guessed electronics lectures would be like 60 years ago. Rectifiers and the things that are presently more esoteric. I’m really glad I’ve found your channel for this, THANK YOU very much!!
This was exactly what was looking for. I have been working on implementing VVR on my amp. Replace the voltage divider with a pot and there you have it. Thanks very much!
This behaved well in sim. Saved me a lot of time. Thanks for the video.
thank you for clear explanation for non electrical engineer tube builders!
Interesting stuff. As a novice tube amp DIY'er its interesting to hear some of these concepts explained. Like others have mentioned it would be even better if you would include how to determine the values of the parts and examples with values. But hey I'm already long happy to see new tube related content :) keep it up!
phalanger1 I may do a more practical in-depth video on specific values. Thanks!
If you enjoy this sort of stuff, go to your local community college and take some classes… maybe get an EETC associates degree.
That would be fun. Wish I had the time!
A calculated example would be the icing on the cake!
use the falstad simulator
chillaxing to watch, i learned a lot
Nice video, explained everything in great detail, we will refer our students to this video. It contains all you need to know to build the regulator. I see some people complaining about a lack of values and calculations. If you can't compute the values for your requirements from the explanations in this video, you should NOT be experimenting with tube amps and power supplies like this.
togethia Thanks, I’m happy you find it useful!
Just look at a schematic the values are there already. If you are into designing your own then read books on how the values are determined. If you change tubes then you need to change values for the most part unless it's a direct substitute. 73
I never had a rectifier tube go bad. Probably because our voltage was only 115 vac from the standard outlet. Filter caps usually went bad probably because of their age. My amplifier had a 5Y3 and the preamplifier stage used a 12AX7 wired for 6.3 volts, the power amp stage was a push pull 6AQ5. Good little amplifier made by admiral. I believe it would still be working if I still had it. I have a voltage reducer for my tube radios. Made by APC the same company that makes UPS for computers. I like to keep everything original they sound better as a tube amp. My biggest project will be restoring my dynaco st70.i bought all new tubes including the rectifier tube. The only thing on the dynaco amp that will be changed is the Selenium rectifier I have the diode that was recommended for that amplifier. 73
That was very enjoyable did I understand what you were saying NO! Yet I will watch it again and again it really was good!
Nice if you to say!
@@Edgarbopp 🙂
This is the best explanation of a main power supply and filtering I have found so far.
Is it possible to use a Mosfet after a rectifier tube or do you have to use a Bridge rectified B+ with a Mosfet only?
The FET filter will work just as well after a tube rectifier. I’ve used it that way in a preamp I designed. 🙂
This is a series Voltage Regulator versus a Shunt Regulator, many prefer a voltage source feeding a Shunt Regulator for Audio Electronics as it is more stable to powers line fluctuations. A Shunt Regulator would have the Drain going to ground and the Source near the incoming voltage..
Nice overview. Thank you for your time and effort!
IMHO the title is a little misleading, though. I would expect calculations if "design" is in the title.
Nevertheless, great video and hoping more of this type come out: subscribed!
Valuable clip, excellence
Awesome 👍👌
For your final example where you suggest the idea of several zeners in series to add some voltage regulation to your circuit, I am recalling the use of tubes such as the VR-150. (Which we used to use.) Also, this brings to mind devices like the NE-2, as well. Just wanted to toss out the idea that one does NOT necessarily require the use of silicon zeners for this purpose.
Definitely agree! There are lots of ways to make a voltage reference. I’ve been using a small depletion mosfet for that lately. Uses way less current than zeners!
When a teenager (world population less than half of today's) I designed the HV regulation for the Geiger tube using a series string of NE-2's to develop the approximate 1400 VDC needed for the tube. (I contacted the physicist responsible for designing that tube and he sold me one, directly, for USD8.) Each NE-2 is a little different. So one needs to measure the stacked result, of course. I hadn't considered your thought about a small depletion mode mosfet, though. Need to think a little about that. :) @@Edgarbopp
On your first drawing you should match the transformer resistance and min rectifier resistance
found on your data sheet about 50-200 ohm add them and put in series a 10 watt or 20 resistors the added value should work very well…put the resistor right after the rect tube this puts a load on the tube and keeping current moving forward and not going back into the transformer and causing it to heat up saving the tube and transformer life many manufacturers did this in the 1960s
also the choke inductor has a resistance from 150-300 ohms as well add these calculations in your circuit design 😀the inductor choke will lower your b+ also using a too large of a capacitor on the first stage could be bad again look at the data sheet of the tube after the choke have fun and put a giant one in but 50mf to 100mf should do plenty
Also you can do a simple 5 stage filter with resistors and capacitors
I personally would use a motor run capacitor for a hvac they are dirt cheap and on Amazon and are oil filled and have no esr and make perfect filter capacitors I have had very low ripple ac about .0031 at -90dbv and a standard ac ripple it will be dead quiet
Nice!!!
The protection diode you added from source to drain in the fet is not required because it already exists as a function of how the fet is created, It is commonly called the 'body diode'...
The zener diode you added as a dual purpose device to current limit the supply and protect the Gate of the fet can actually cause the fet gate to go overvoltage in the case of the input cap shorting out while the output cap is still fully charged.
I strongly recommend using a bipolar transistor to sense the voltage across the current sense resistor with a series diode (1N4007) between the collector of the bipolar transistor and the gate of the fet which would allow the transistor to pull down the gate voltage to current limit, but not allow any current flow TO the gate when the output cap is charged when the input cap voltage rapidly drops to zero.
You also need to add a reverse biased HV diode from the gate to the drain of the fet because of the reverse voltage pulse that can be developed across the resistor between the capacity multiplier input and the charged capacitor from the gate to ground.
There are other ways of providing similar protections, but the one I described fits well into your design and provides a more accurate threshold of current limiting.
I definitely agree! I’m still just learning and since making this video I’ve actually used a BJT across a sense resistor just like you described!
Some FETs do have reverse voltage protection built in but some still don’t so I tend to leave it in as a extra uf4007 costs basically nothing.
Thanks for your thoughtful comment!
@@Edgarbopp
Thank you for responding.
Yes there are different types of fets, namely jfets and mosfets, adding a little more confusion, both types are available in both enhanced and depletion models.
There is NO physical diode inside a mosfet, the diode is the result of the process used to make it! Therefore I cannot imagine you finding a mosfet anywhere that does not have a body diode,, And you were using an enhancement mode n-channel mosfet in your schematic.
Please also spend a little more time considering the other modifications I recommended. I have an illness called MG, and it took a very long time to type that comment because of problems controlling my muscle movement..(the shakes, but not old age shakes).
There is no doubt that your design will work, I was only trying to help you improve your protection systems.....
@@Dennis-mq6or thanks for taking the time! I’m sorry you’re struggling with MG!! I’ll reread your comment carefully.
Hi Doug,
I really enjoyed this and liked the way you walk through the circuit from something basic that is understandable, and slowly make it more complex. Each time explaining really intuitively the improvement.
I did have one question. I really like the way old amps like Plexi etc briefly sag when hit hard. And I also really dislike hum! Do you think we can get the best of both worlds with a design like this?
For example something quite saggy, momentary voltage drops across the board, and also low ripple? Just wondering if you have played around with that?
This is absolutely possible. You could for example use a SS rectifier and FET filter then follow it with a traditional series tube rectifier smallish cap and choke. Should have identical sag/power compression characteristics with none of the hum.
A less complicated option would just be adding series resistance after the filter until you get the voltage drop you prefer when the amp is driven hard. Keep in mind this only really works with push pull class AB amps as class A amps draw constant current even when driven hard.
i Like your psu design.
I wanna use/have 2 x 15 Volt (1.5 Amp) transformers back to back, to get around 210 V ac ( a bit of lost from 220 V ac). after i rectify that i have like 290 V dc.
the circuit i wanna build has 3 12AX7 an 1 6AQ5A tube, both have a max operating voltage of 250 V.
What output voltage of your psu would you choose? the schematics says 275 V dc.
i saw your example circuits and taking "p7", what resistors, zener diode, Mosfet, capacitor (appart from the 4700uF, next to the rectifier) would i take?
New to the channel and am very impressed! So have you given any more thought to a real live with values example video on this? And if I may ask....what are those depletion type Moffetts you've spoke about in the other comments? Thanks!
LND150 for a low current voltage reference. It comes in TO92 as well as other small packages.
Another question if I may: Because of horror stories I've heard about when using modern diodes in place of the old rectifer tubes and problems they can cause is that the reason for that RC time you mention ranging in the 15-30 seconds, is that to mimic the time the old tube rectifiers would also give .....as in, to help illuminate current surge and time for the tube to warm up before seeing full voltage on their plates ?
Can I use the tube power supply with a solid state mini amp I just built?
Excellent video. And passive components for high voltage? Suppose I am building a transmitter with an RF vacuum tube needing 800Vdc plate voltage. What capacitors and inductor can I use? And where do I get them? Oops. I did not watch the whole video. At 16:00 into the video the inductor is replaced with an N channel MOSFET. I wonder if I could use P channel for negative supply?
I should make a video about that. There are some pretty simple ways of achieving 800v using only 450v rated capacitors.
Nice job. you're Very talented at making videos. stay frosty x3
Thanks!!
Tube rectifier will be great for your tube amp.
Bravo bravo bravo
Great Video found it really useful, would you mind explaining further the job of the resistor on the output after the source of the Mosfet.
This is a current limiting resistance. If the drop across this is excessive and if you’ve got the correct Zener value this will set the max current of the mosfet. This may save the circuit in case of temporary short.
@@Edgarbopp thansk 👌
Hiya probably a very nube question, but coud you get away with lower voltage capacitors after the voltage divider on the gate, as the voltage cannot be to high so as not to exceed the gate voltage.?
@@phenome4973 sure, if the voltage divider is such that the gate voltage is low enough.
Any reason that the source follower circuit won't work with a full wave rectifier (not a bridge)?
It will work fine with either. Even a half wave.
Wuld you please make those old wacum tube baterry charger circuet diagram? Im surching for some thing similar that can by used for magnet alternators I like wacum tubes but dont know how the automatical current cutt out circuet shuld look like with wacum tubes is it posibel useing those old external voltage regulators or those old external regulators for dc generators for cars with wacum tubes. Honestli told Im indead afraid to use it in car so was curriuse can that by some sort of cloosed loop system?
Why not put an inductor in-line before the final mosfet circuit? Get the best of both worlds? Would be interesting to see how much ripple makes it to the final circuit under load.
Nice vid, thanks for making it. I stumbled in here looking for a regulated DC power supply made with just vacuum tubes. Yeah, I know, I'm a glutton for punishment :)
You could put a inductor in series. Not sure how much it would help. I do need to do another video on PS design I should include tube regulation!
What about negative feedback to make it regulated?
Really nice job and great explaination of how things work. I see the link to Circuit examples is broken . Do you still have circuit examples avaiable ? I'm building a power supply for a BC-348. It is an HF receiver from a B17 Bomber. I think the regulated version fits the bill to replace the Dynamotor.
I’ve fixed the link!
@@Edgarbopp Thank You ! I have the supply simulated with pspice and seems to work fine so it is on to building it. I chose the regulated version with a 214V zener value to yeild a 210 volt supply from a 350 V source. Thanks again.
Good explanation of the different supply options. How does this final circuit differ from "capacitance multiplier? To my untrained eye it does not. I'll try the last one out. Is it necessary to filter it further with an inductor or can I pass it straight to plate?
The final circuit has a voltage reference created by a shunt regulator. In this case it’s a resistance and a Zener. That locks the MOSFET gate at a specific voltage. So it’s more like a regulator than a capacitance multiplier at that point. The output may feed the output tube plates directly. It will be very well filtered
@@Edgarbopp Thanks for explanation. Can I use high voltage NPN silicon transistor instead of a MOSFET? There are also 150v and 200v zeners available - how do I set the voltage for the transistor with it? I need like 250V in the output with up to 100 mA of current.
@@MarkErikEE a BJT can be used but the base will require current unlike a mosfet.
To generate your reference voltage you will need to create a shunt regulator with a resistor and a Zener or a string of zeners. You will need to supply a minimum current to the Zener for it to drop its specified voltage. That should be in the data sheet. Chose the the resistance by understanding that the remaining input voltage that is not dropped by the zeners will be dropped by the resistance.
Hi Doug - can you tell us how to compute the components value for a 320V output? 250VAC PT? Many thanks!
If the HV secondary is 250v a quick way to guess the voltage after bridge rectification and filter cap is to just multiply by 1.4x that will get you in the ballpark.
@@Edgarbopp Indeed the secondary is 250 which yields about 350 after the first cap. What about the drop of voltage on the MOS circuit? How an I compute the components values? Many thanks!
Would love your comments on how tube rectification SOUNDS.
Tube rectifiers have a “sound” only under very specific circumstances. They have a highish internal resistance and so will drop voltage depending on the current. This means that when used with a class AB output stage the voltage drop will increase as the stage is pushed harder. This is called voltage sag. In instrument amps this can cause a combination of slow attack and release compression as well as additional distortion. Some musicians like this. However if SS diodes are used instead of a rectifier you can simply add a power resistor in series with the HV supply to mimic the internal resistance of a tube rectifier.
if a thermistor was put in the power supply ahead of the transformer wouldn't it help to protect the rectifier tube
Certainly interesting, although I have my doubts.
This looks more complicated as it needs to be - in my reading of tube systems schematics I've never seen anything similar. In a lot of modern tube power supplies LC filter is simply replaced with RC filter (respectfully series of them). Also, you have said it wrong, most of the times OT primary is connected directly after the rectifier (on very first cap) and after the choke comes the second node, going to the screens. But surely a similar circuit would be quite nice to provide a bias supply (even eliminating the need for a separate bias winding). I think you are also wrong about the CT PT with two diodes being in any way inferior to the bridge rectified configuration, I suspect it is much easier on PT and eliminates half of the rectifiers needed.
This for example is a quite modern Slodano 100 PSU www.trioda.com/images/gallery/soldano_100/001.gif
Back when rectifiers were all tubes, using a center-tapped transformer was the cheapest way to go, as tubes were pretty expensive. These days, rectifier diodes are cheap, you can buy a 2A bridge rectifier for under 50 cents, and copper for transformer windings is expensive, so cutting the amount of copper you need in half is a definite plus.
For the same reason, big filter chokes are expensive, so usually an RC pi filter is done instead of an LC pi filter.
With the Zener regulator you show at the end, if wall voltage drops and the zener drops out, does the circuit go back to being a capacitance multiplier?
Yes!
@@Edgarbopp Well this is interesting. That means one should probably keep the higher value cap in place. It also means you might as well always implement the zener regulator. I wonder which of the two circuits, regulator or CM lowers noise the most?
@@rhclark6530 the zener regulator and CM are actually very similar in terms of ripple. I prefer the CM because you waste less power. With a regulator you need sufficient V drop so that the circuit won’t drop out under any circumstances. A more sophisticated regulator with a error amp and feedback network will be better then the zener regulator. Check out my video on the Maida regulator for more on that.
@@Edgarbopp I see your point. Also I guess one could cascade two CMs for potentially insane ripple reduction in phono stage or headphone amp potentially. Yeah, the Maida video is excellent. Best explanation of the circuit I have seen.
@@rhclark6530 my current pre uses a reg into a pair of CM.
See ELECTRONICS FROM THE GROUND UP by Ronald Quan, P 499
14:20 You messed up the diode bridge description - opposite sides of the bridge diamond are conducting on each half cycle.
You’re right. I totally messed that part up.
Aren't you significantly raising the power supply impedance with that current resistor at the output?
The output cap dominates the output impedance of the circuit.
@@Edgarbopp Ah. I have yet to find a good explanation anywhere of why power supply output impedance in audio power supplies matters and what we should do about it when designing? Do you know of any? Maybe you could do a video? ; )
@@rhclark6530 I have a video on the output impedance of the amplifier but the output impedance of the power supply is a good topic idea. Typically it’s important for lowering crosstalk between channels, and providing adequate current in class AB amps when they transition to class AB from A.
@@Edgarbopp Interesting. I hope you do a video. One thought is that local bypass caps at each tube or each channel might help with shared power supply crosstalk. BTW I think you were implying earlier was that irrespective of what comes before it you want a big cap at the output of the power supply to keep the impedance of the supply low? If that's true what's the point of diminishing returns. 47uf or 470uf?
@@rhclark6530 the size of the cap needs to bypass all audio frequencies at the impedance of the circuit. So the size necessary is different for different applications. Though as a general rule, I find 100uf to be a good place to start for power amps in the 10-20wpc range.
I built an LM386 small amp, can I use a tube power supply instead of a battery?
What would you want to achieve? Bulky and potentially dangerous thingy, and a lot of problems getting the voltage right for LM386?
@@JanicekTrnecka Is it wrong to experiment?I know the risks of working with tubes.
Definitely not, but I dont see much sense and reasonability in powering LM386 with tubes. I would rather build "battery powered tube amp" instead of probbably grid powered tube power supply to operate LM386...
Please dont feel dissed... I have built a lot of "unusual and pretty weird" circuits too ;)
@@JanicekTrnecka I'm sorry for my reply, we don't see face expressions in comments to understand the real intentions of them. I'm retaking what I learned years ago about electronics, I totally forgot what I learned. Cheers!
@14:26 directions of returning currents are wrong. They are returning at the other end of winding than they are entering the circuit.
I got a bit confused there. Thanks.
Just some silly idea is that inductor there by any chance replacebel for exampel dc commitator or armature to by preciseli what wuld the result by dc generators use wave and lap windings expeciali Im refering to Amplidyne and Metadyne generators . So Im cinde of interested how those old gasoline electric trains operated exactli no one hase rewive video about them on youtube . So wacum tubes are reali cool devices
Mister ap ad yg pakek BHS indonesia
cap choke filter
Yes, nice, but fairly useless, so far. Not one calculation. How do you build power supplies with no calcs, by "feel"?
Didn't want to get into the weeds with this video. More of a zoomed out view.
Dude, do everyone (including yourself) a favor and write a script. Its really hard to watch and understand your videos with all the "uuhh" and "uuummm" in that monotone. The content is cool, the execution is rough.
Please feel free to not consume my content.
hi sir my name is mohammad..in the past we talk together..but sir i have a problem,, i have a otl circuit with 6080 vaccum tube ..this is my question :: who must i calculat the anod resistor??? because i have a power suply with 230volt on output but i need to 170volt on the anod!!!!!! now i need know how can i calculat anod resistor????? and i dont want change anod current just i want change the voltage .....if you need to know more detail,, you just tell me and i will send for you my email ...... pleas lead me and i am realy thankyou.......