Built Renegadrians 555 SMPS. Only pulling 155V

[illuminatiNPS]

So I built the original rick holt 555 smps. The original was moved to vero by Renegadrian. I read that its supposed to put out like 200v unloaded but I am stuck at 150v. I am feeding it with 12v dc. I have already to replaced the irf740 and inductor, same results. Anybody mess with these. here is a link to the vero


http://www.aronnelson.com/gallery/main.php/v/Renegadrian/555+smps+vero.gif.html

[M23Bomber]

 

Hey man,

Make sure the PSU is 12V DC 1 AMP, swap the 2,2uF to 4,7uF. And also dont use the CMOS 555, make sure you are using the Bipolar 555 them you can get better results.

Regards,
M.

[thomasha]

Hi,
I tried this smps with a 6v 800mA wallwart it achieved 250V with load (ef184 preamp).

I would say your problem is the diode, is it a ultrafast diode?

It needs to be UF400x and not the 1N400x, because of the switching frequency.

If you're using an high voltage ultra fast diode, my second guess would be a problem with the transistor bc547 or the resistances in the voltage control part. Check if this transistor isn't inverted.

Check the resistor values, the 220k and 470r.

Cheers,
Thomas

[illuminatiNPS]

Hey guys. I am using a 12v 1 amp psu.. The diode I am using is a uf4004. It is a bipolar 555. I will check the orientation of the bc547. I figured the layout shows the curved part of the trans facing left so that is what I did. when loaded with a 220k resistor it drops to 130v. Also the 1k trim pot is only shifting 3 or 4v in either direction. though it should be more.

[sajy_ho]

Does the mosfet get too hot? I found some of IRF740s getting too hot in this circuit, while others from another manufactures stay dead cold!
Although the 555 chip is not much layout sensitive, but a very bad layout could be the source of problem too.
If you're familiar with pcb etching, you can use this layout:https://db.tt/NqrEqe1L
It's a high efficacy layout and I used it to power PRR's "Third Watter" with success.

[M23Bomber]

Hey all,

Do you know if the 555 SMPS can power up a 12AU7 in push pull configuraration with a  12ax7 preamp? Basically the FIREFLY with a SMPS and only 2  tubes, also know as the SUPERFLY with easy to get tubes.

Knowing that the efficiency of the 555 SMPS is arround 30% to 35% , meaning, from my mental math , 15 to 19 mA at 220V. It would kind run,right?

Regards,
M.

[sajy_ho]

Hey all,

Do you know if the 555 SMPS can power up a 12AU7 in push pull configuraration with a  12ax7 preamp? Basically the FIREFLY with a SMPS and only 2  tubes, also know as the SUPERFLY with easy to get tubes.

Knowing that the efficiency of the 555 SMPS is arround 30% to 35% , meaning, from my mental math , 15 to 19 mA at 220V. It would kind run,right?

Regards,
M.

I did some tests on three SMPS's using NE555, MC34063 and MAX1771 a while back and fortunately I've still got the results in a notebook.

I used the same layout that I posted above for the 555 smps, for max1771 I used Nick Desmith's layout. Then I messured the highest output voltage that the tested smps can put out using resistive load and also measured input current and efficacy too.

Here are the results:
555 SMPS:
Max. 12mA @ 250V with 53% efficacy
Max. 15mA @ 220V with 62% efficacy
Max. 19.9mA @ 190V with 72% efficacy

Mc34063:
Max 11mA @ 205V with 47% efficacy
Max 16mA @ 155V with 39% efficacy

Max1771:
Max. 19.5mA @ 280V with 71% efficacy
Max. 28.1mA @ 220V with 73% efficacy
Max. 34mA @ 204V with 79% efficacy


The Firefly needs around 18mA @ 265V, so I think the only way you can power it with a smps would be the Max1771 smps; but if you bias the output stage very cool you can use 555 smps too...

[M23Bomber]

Hey sajy_ho , your testing and that notepad have made my day Thank you very much, for sharing your results

For example in this schematic:



So your advice is to increase the  R13 from the 440ohm , to a higher value ?

The way I see it , a resistor and potentiometer in series would do the trick,  such as 220ohm/2W resistor and 500ohm/2W potentiometer)

Regards,
M.

[thomasha]

Hi,
I would just reduce the b+ voltage.

And put a heat sink on the irf740.

If you're go as low as 220v you can even change it for a lower voltage fet, which will give you less heat, like the IRF644.

The inductor should also be a low esr one.

Cheers,
Thomas

[sajy_ho]

Hey sajy_ho , your testing and that notepad have made my day Thank you very much, for sharing your results

For example in this schematic:



So your advice is to increase the  R13 from the 440ohm , to a higher value ?

The way I see it , a resistor and potentiometer in series would do the trick,  such as 220ohm/2W resistor and 500ohm/2W potentiometer)

Regards,
M.

Yes I meant to increase the 440R cathode resistor. using a potentiometer would be a good idea in a case that you find a better smps in future and you can adjust everything easily.

As thomasha said a better idea would be to reduce the B+ to around 220V and use stock 440R resistor, also using a lower voltage\higher current mosfet means more available output current and less heat.

[gtudoran]

For this kind  of SMPS the layout is critical (as for any SMPS for that matter). Running it on a perf-board may not be the best idea at all (when i've made the first prototype of my SMPS like 4-5 years ago i've run it on the breadboard ... nothing over 180V).
I'm using a layout made by me - use it in all my tube builds (from distortion pedals to small tube amps - you can look for the SFA amp posted here) - but is double sided, low voltage on top / high voltage on bottom, short traces, the output pin of the 555 (pin 3) tied with a short trace to the gate pin of the FET (pin 1) / thick traces from the VCC to the inductor / ground plane all over the surface / at least 1.8A inductor / an original FET made by a known manufacturer (Fairchild / ST etc.)  / low ESR output cap - don't have one use a 4.7uF (do not exceed 10uF for the first cap) (at least 400v) in parallel with 100n ceramic (2KV ceramic should be fine) - and i'm using over 50KHz switching freq. - also i recommend you to use a known manufacturer for the inductor.
With this configuration i've managed to get over 350V (all my pedals are running at least 340V under load - 2x12AX7)- but ... you will have to comply with the design rules for a SMPS
Video: https://www.facebook.com/dextersworkshop.ro/videos/vb.100006730863050/1622922957942051/?type=3&theater

Regards,
DeX

[Renegadrian]

Hope you did get past your problem in the meanwhile! Anyway I got one built with my layout attached to a 12ax7 and it gives circa 300V maxed, 150V with trimmer in half position. It really depends on the components you use, diode and inductor are critical.

[Rob Strand]

Firstly I think there is a minor issue with the layout. R6 and C2 should connect to the anode of D1.
These parts protect the MOSFET to some degree.
See the schematic,
http://i210.photobucket.com/albums/bb292/frequencycentral/555SMPSNegative33.jpg

I don't think that's the issue though.

You will see what you are seeing if the inductor is saturating.
To get a higher output voltage the MOSFET on-time needs to be longer.
That means the inductor current will ramp up more.  However if the
inductor saturates the current will go through the roof and can pull
down the effective input voltage due to IR losses in the inductor.

For the 555 design the off time is fixed.  In that design it's 15uS.
To get 200V out (ignoring losses) that means the on-time needs to be (200/12) * 15us = 250uS

With a 100uH inductor at 12V on for 250uS I get a peak current of,
Ipk = 250us * 12V / 100uH = 30A !!!!!
That points to a design issue.

Anyone agree or disagree with this?  (I could have made an error)

The minimum pulse width from a 555 is somewhere in 3 to 5us region.
Say we make it 5uS to be safe that means you can decrease the 2n2 timing cap
by a factor of 3, say 680pF, which will then decrease the peak current to about 10A.
The other way to go is to use a larger inductor value.   I suspect you need a physically larger inductor.


This only happens at start-up.  It should back-off the pulse width once you reach voltage.

I suspect the inductor losses are the cause  ie. DC resistance too high.

[GibsonGM]

I just built this.  I used the schema you link to, Rob, and also the "Nixie Supply" PDF that's all over the internet.

Variations of this particular SMPS have been spotted floating around, missing at minimum the 56k resistor from + to pin 5 & collector of BC547!  That is critical and will greatly limit the output.    I have this SMPS on perf right now, feeding a GTFO with just jumpers, and it's very quiet...my inductor is a garden variety from the junk box (not sure of its value, even!!), and I used IRF 840.

My thought is that the diode and non-CMOS 555 are critical, and also - NOT trying to crank it up too high...250V perhaps?  This thing makes SPIKES (above audible frequencies) that can take out filter caps!   So, I run mine at 240V with good results.    If I did another, I'd try harder to find the correct inductor...

[Rob Strand]

This is what I was looking for:

https://www.researchgate.net/figure/26899555_fig7_Fig-12-Output-voltage-vs-duty-cycle-boost-converter-with-copper-loss

When you study SMPS design they often show this graph for a boost converter.   What it shows is the inductor DC resistance limits the maximum output voltage achievable in a boost converter.  No adjustment of the duty cycle will allow the voltage to get higher.    In reality it's not just the DC resistance that contributes to this limit but all sources of power loss.  It's just that the inductor is often the culprit.    It also depends on the load, in fact the link shows the ratio of load resistance to inductor resistance.

To get 200V I suspect you will need a DCR less than 1ohm.   A 100uH ferrite drum inductor 8mm diam x 10mm high would have say 0.5ohm (Isat ~ 1.2A).   That might do it.  Where as a physically smaller 2 ohm or more inductor might not make the 200V.    Please don't treat these as exact values.  I'm just ball-parking  things.


I

[Renegadrian]

Firstly I think there is a minor issue with the layout. R6 and C2 should connect to the anode of D1.
These parts protect the MOSFET to some degree.
See the schematic,
http://i210.photobucket.com/albums/bb292/frequencycentral/555SMPSNegative33.jpg

Gonna update the layout, thank you!
I put it into an alembic F2b but the wall wart is too noisy, it can be used at minimum (circa 90V) but the more I turn the trimmer the more noise and whine is strong - tried a 9V 300mA WW and I got no noise, got good voltages but obviously the heater voltage is too low (circa 8V) even if it works in a good way.

[GibsonGM]

Noise & whine:   Is that a symptom, then, Adriano?  The thing switches at over 40kHz....shouldn't it be quiet?  The input cap should get rid of a lot of noise, shouldn't it?    I think it has to do with what inductor you use. I got lucky, it seems.     They *seem* to be hard to find (barrel type, low DCR).  The one that was sourced a long time ago isn't available anymore.   I got mine in a pack from RS like 15 years ago, it was just in the junk box!

I checked on a scope and found some pretty high voltage spikes occurring, so I rated my preamp caps at 450V, and only run this at 225 or so.   Just in case.  I think I did pop a filter cap the first time thru! 

[Rob Strand]

The thing switches at over 40kHz....shouldn't it be quiet?  The input cap should get rid of a lot of noise, shouldn't it?    I think it has to do with what inductor you use.

The simple SMPS's inherently vary the frequency.   You get can also get weird behaviours with light loads.   The minimum on and off times of the 555 are limited so it might turn out that with light loads (and less taxing conditions of lower o/p voltage) the control transistor shuts the 555 off and you get an erratic o/p frequency.

I suspect playing with the inductor value may help.  If you put a crappy high resistance inductor in there it might help too but that just makes the converter less efficient.  It keeps out of the weird areas of operation by the shear fact it needs to work harder to get over its own losses - not really a good solution!.

The modern SMPS controllers have schemes to help the cause (burst modes, pulse skipping and some monitor the inductor current.).   There's a bit more going on than the simple 555 ckt.

[GibsonGM]

I guess I lucked out then, with the cheapo inductor I found.  Mine turned out quieter than a brand new amp idling, and I didn't worry much about layout!

But that was after I had to do 3 or 4 things to get it to work!    If I did one again, I'd want to use a 'tried and true' inductor.

Just FYI, This thing draws about 40mA from the 13.5V/1A wart I'm using, IIRC, and I suspect the 4 triode sections of the GTFO aren't drawing more than a few mA.   It is drawing the voltage down so little on that wart that I used a dropping resistor for my heaters, which are also operating from that wart.