Don't know if this has been mentioned before, but when you speak to Toroidy mention that you need the transformer for audio; this will get you a bigger core and a thicker wire used for the secondaries. Gets more pricey as well. Not sure if they advertise it on their homepage though.
They are specific on that matter:
Standard grade (casual)
Audio grade (potted core)
Supreme audiograde (potted in a glossy (hopefully mumetal) shield)
They probably have different materials for the windings as well but don’t know.
Price go up of course (the first item is a 20VA, prices are 60 - 100 - 299 zlotys, just to give you a perspective)
(And no, I am not affiliated, just like them audio-grade trannies 🙂 )
Not associated to Toroidy, but the Supreme range is as close as you're going to get to DIY hifi p*rn for power supplies.
Definitely worth the additional money for a build you don't plan to mess about with often.
I used a supreme audio 400w for my f4 after the 500w sedlbauer I had was humming. Everytime I don't get a torroudy I generally regret it. Mine had a shield wire which bolts the the chassis.
I was tricked by the Polish Zloti to USD conversion lol. The prices look way smaller after conversion so I decided to splurge a bit.
lol - I did the same initially to GBP and I've been to Poland multiple times 🙄.
They're actually really reasonably priced, let us know what you think - if you end up going for them.
For F3
Would it be possible to get a power supply pcb set with markings for 2 +ve voltages, for example for the First Watt F3? If not will use the current pcb set (one +ve power supply will have incorrect markings).
SMathews
Would it be possible to get a power supply pcb set with markings for 2 +ve voltages, for example for the First Watt F3? If not will use the current pcb set (one +ve power supply will have incorrect markings).
SMathews
Use the standard PCB. It can be configured as such. Yes, the markings on the output will be a little imprecise, but still understandable.
Is more capacitance always better?
If I can afford more capacitance is there any advantage or downside? I'm thinking of going for a total of 198000 uF instead of 132000 uF of a board that has room for 6 capacitors (from Class D Audio)
Bruce
If I can afford more capacitance is there any advantage or downside? I'm thinking of going for a total of 198000 uF instead of 132000 uF of a board that has room for 6 capacitors (from Class D Audio)
Bruce
Hi All,
I built a V3 PSU, 24vDC rail, 60,000uf / rail and 4ea .47ohm resistors / rail.
I measured the PSU ripple loaded with a ACA v1.8 at idle, it measures 296mV.
I'd like to get the ripple down a bit more, would adding some of the optional resistors help or maybe look at changing the capacitors out?
I built a V3 PSU, 24vDC rail, 60,000uf / rail and 4ea .47ohm resistors / rail.
I measured the PSU ripple loaded with a ACA v1.8 at idle, it measures 296mV.
I'd like to get the ripple down a bit more, would adding some of the optional resistors help or maybe look at changing the capacitors out?
Last edited:
296mV of ripple seems like a lot for a CRC of 30,000uF - 0.12R - 30,000uF.
Did you set up the power supply for one rail per channel? So that each channel has its own CRC, for 120,000uF total for two channels?
And how did you measure the ripple? Oscilloscope? Any noise/hum at amplifier output?
More information/details/pictures would be helpful.
Did you set up the power supply for one rail per channel? So that each channel has its own CRC, for 120,000uF total for two channels?
And how did you measure the ripple? Oscilloscope? Any noise/hum at amplifier output?
More information/details/pictures would be helpful.
Adding the optional resistors will reduce the total resistance (because you are adding them in parallel) which then reduces the overall filtering, which will have the effect of slightly increasing ripple.
I'm using this PSU to power an ACA v1.8 until I build the F5.
Transfo is Antek 5218 (240vAC in)
PSU is setup for -24, 0, +24 vDC.
This test was on the +24 rail, so test is against 60,000uf of capacitance only.
Yes I do have very slight hum at speaker if pre is connected to ACA, if I disconnect the RCA inputs the hum goes and its dead silent.
I know I have a ground loop and quite sure its within the ACA and will address that next once I get this ripple down a bit.
Did you set up the power supply for one rail per channel? So that each channel has its own CRC, for 120,000uF total for two channels? Yes correct
Sorry for the crappy scope pic, damned if I can find an app that will display Hantek exported screen shots!
Transfo is Antek 5218 (240vAC in)
PSU is setup for -24, 0, +24 vDC.
This test was on the +24 rail, so test is against 60,000uf of capacitance only.
Yes I do have very slight hum at speaker if pre is connected to ACA, if I disconnect the RCA inputs the hum goes and its dead silent.
I know I have a ground loop and quite sure its within the ACA and will address that next once I get this ripple down a bit.
Did you set up the power supply for one rail per channel? So that each channel has its own CRC, for 120,000uF total for two channels? Yes correct
Sorry for the crappy scope pic, damned if I can find an app that will display Hantek exported screen shots!
Attachments
Last edited:
I see that you are only using one rail in your picture. Are you powering both channels with the one rail, drawing 2.9A or only one channel? I had assumed one channel per rail, so 1.45A per rail. The current draw will affect ripple. The higher the currrent, the higher the ripple.
Also AC frequency will affect ripple. 50Hz has higher ripple than 60Hz. I had mistakenly assumed 60Hz.
So with 2.9A and 50Hz, your measured ripple is closer to what I would expect, although still higher than the 200mV that I got from a PSU Designer II simulation. But it is in the ballpark. If you are powering only one channel, drawing 1.45A, then the simulation showed 100mV ripple.
As Jim mentioned, adding resistors in parallel will reduce the R and increase ripple. Increasing the R will decrease ripple.
You have noticeable hum only when a preamp is connected, and no hum when no preamp is connected. That would indicate that power supply ripple is not the cause of the hum, but some sort of grounding issue is the cause.
Also AC frequency will affect ripple. 50Hz has higher ripple than 60Hz. I had mistakenly assumed 60Hz.
So with 2.9A and 50Hz, your measured ripple is closer to what I would expect, although still higher than the 200mV that I got from a PSU Designer II simulation. But it is in the ballpark. If you are powering only one channel, drawing 1.45A, then the simulation showed 100mV ripple.
As Jim mentioned, adding resistors in parallel will reduce the R and increase ripple. Increasing the R will decrease ripple.
You have noticeable hum only when a preamp is connected, and no hum when no preamp is connected. That would indicate that power supply ripple is not the cause of the hum, but some sort of grounding issue is the cause.
Morning Ben Mah,
thanks for the response. Yes powering both channels from the one rail.
This PSU is ultimatly intended for a F5 but I just wanted to see how it worked with the ACA.
I do plan on building a PSU for this ACA and get rid of the SMPS, there are noticeable benefits with this PSU on the ACA.. dynamics & stringed intrument resolve
Given its drawing ~2.9A from a single rail and you say the ripple is within the ball park I'll tick this off and move onto the hum.
I'm 90% sure the hum is due to how the ACA is wired (as per v1.8 schematic).
Currently both channel PCB earths tie back to a common earth - star.
I plan to change this to a series earth, whereby a single wire from common earth goes to one PCB then jumper from there over to the other PCB.
If anyone has any other suggestions I'm all ears!!
Thanks
thanks for the response. Yes powering both channels from the one rail.
This PSU is ultimatly intended for a F5 but I just wanted to see how it worked with the ACA.
I do plan on building a PSU for this ACA and get rid of the SMPS, there are noticeable benefits with this PSU on the ACA.. dynamics & stringed intrument resolve
Given its drawing ~2.9A from a single rail and you say the ripple is within the ball park I'll tick this off and move onto the hum.
I'm 90% sure the hum is due to how the ACA is wired (as per v1.8 schematic).
Currently both channel PCB earths tie back to a common earth - star.
I plan to change this to a series earth, whereby a single wire from common earth goes to one PCB then jumper from there over to the other PCB.
If anyone has any other suggestions I'm all ears!!
Thanks
Last edited: