Tangent Young-Jung power supply

[grenert]

I see that Tangent has been working on a new power supply that appears to be a replacement for STEPS:
Young-Jung Power Supply

I know he had some PCBs available for a while. Have any of you tried it out yet?

[rds]

I'm building one tomorrow. I'll post some pics.

[TzeYang]

One crucial question would be how much capacitance will this regulator drive stably?

[n_maher]

Did he finally put it into production? The design has been around for years now and as far as I know was never finished.

I've had one going for ~2yrs now in my Millett Hybrid and also built one for my PPA years back which is still in service. One important thing rds, do yourself a favor and modify the pcb to allow you to bolt down the two large heat sinks. I had one snap off in transit. For one of the sinks it's pretty easy, for the other you can only get a fastener on one side but that's still better than nothing. And if you have any questions just ask, I may or may not remember the answer but I can always just look at the board that I still have.

[rds]

if you have any questions just ask, I may or may not remember the answer but I can always just look at the board that I still have.

Thanks Nate - I appreciate that.

[tangent]

Originally Posted by TzeYang

how much capacitance will this regulator drive stably?

I don't know where the limit is, but I have one powering a PPA that I use daily, which has never had a fault. I probably put nearly 2000 uF onto its rails.

Originally Posted by n_maher

modify the pcb to allow you to bolt down the heat two large heat sinks

Consider it done.

There will be at least one more prototype run of this board, so now's the time to make requests.

[n_maher]

Originally Posted by tangent

There will be at least one more prototype run of this board, so now's the time to make requests.

I believe that I mentioned this way back when I gave you my build notes but have you ruled out a Vout trimpot simliar to the STEPS design? Personally, I have no issue with the resistor setup the way that it is on the current pcb but the tweakers out there might not be happy until they can dial in exactly 24V out (or whatever).

One more thing for rds: watch out for the on board fuse location. I found that the spacing of the mounting points was a bit too close for the typical fuses. It might be easier to just use an off-board fuse if you have the same version pcb that I had. I know tangent was going to address it in future versions.

[Mazuki]

Can you space out the I/Os so that solderless terminal blocks can be installed?

[JamesL]

Originally Posted by Tangent

I made a small prototype run of YJPSes for early adopters, and am waiting to hear reports about how those are working. In the meantime, I'm working on an improved version. This will probably take a few months to result in boards en masse.

I wonder if the 'improved' version suggests changes to the board layout, or the schematics. To be honest, I wasn't too thrilled about the steps, but the improved regulator on the yjps looks really promising.

[rds]

I believe that I mentioned this way back when I gave you my build notes but have you ruled out a Vout trimpot simliar to the STEPS design?

It looks like you can squeeze in a 1k trimpot at r11. I think that would do the trick

[tangent]

Originally Posted by n_maher

I believe that I mentioned this way back when I gave you my build notes but have you ruled out a Vout trimpot simliar to the STEPS design?

Yeah, I wasn't wild about it then, and I'm not wild about it now.

I think it's because it goes against my concept of a precision DC power supply. Making it adjustable will make the feedback loop larger, and it implies that the output won't be exactly as-designed, within component tolerances. I guess I could talk myself into a very small trim pot that adds to one of the two fixed Rs for compensating for those component tolerances, but then you risk drift. Ech...

My opinion on this isn't set in stone...this's just what I think right now.

I found that the spacing of the mounting points was a bit too close for the typical fuses.

I'll look at it.

It might be easier to just use an off-board fuse if you have the same version pcb that I had. I know tangent was going to address it in future versions.

The ones that went out recently were v1.0. Maybe you could crack the case on yours to see if it was, perhaps, 0.5 or even 0.4?

Originally Posted by Mazuki

Can you space out the I/Os so that solderless terminal blocks can be installed?

Already on the Wishlist, definitely to be done in the next version.

While on the topic, here's the current Wishlist:

• Move secondary-side parts farther away from primary, 8mm minimum distance
• Make it possible to parallel transformer outputs. (Not helpful with 15 or 25 VA, but will let you use 5-10 VA transformers, when lower supply voltages are allowed.)
• Move C8 out from between output pads, to allow use of terminal block
• Add extra GND pad(s)
• Add silkscreen labels to board to make use of unencapsulated toroids easier
• Bigger X and Y cap positions than STEPS
• Add holes for heat sink mounting screws
• Double-check fuse clip distance

Any type in particular you want to use? I had in mind Molex KK's.

Originally Posted by JamesL

I wonder if the 'improved' version suggests changes to the board layout, or the schematics.

Anything concrete?

[rds]

Do the schematic R values give a 30 Volt output?

EDIT - so it looks like the stock values of r11 = 1k and r10 = 1.2k give 15 volts. It also looks like the ad825 can't output more than 30 volts. r11 = 1k and r10 = 3.3k will do 29.8volts. Bad idea?

[tangent]

D3 zeners in normal operation, so it should work fine when set at 30V.

[TzeYang]

there's something i don't understand about the schematic.

Where does the collectors of Q3 and Q4 connect to? The positive rail?

[tangent]

Heh, heh. They don't connect anywhere.

I could tell you what they're doing, but this is a good EE puzzle. (Hint: AoE, page 62.)