There are group buys from time to time and some leftovers. Not assembled, just boards and some key components. PM Tea-Bag to ask if he has any.
hi everybody, hi salas,
i have two ultrabibs to connect to a soekris dac board. in terms of cooling i would fit the regulaturs would near to the side wall of the case.
but i also heard, that its best for shunt regulators to be as close as possible to the dac-board to keep the cable from shunt to dac as short as possible.
if i am going to cool the parts with the side wall of the case, the cable would be at least 13-15cm long. too long?
(i do this only as a hobby, no technical background at all, so sorry for my perhaps very basic question.)
thanks for ideas, best horst
i have two ultrabibs to connect to a soekris dac board. in terms of cooling i would fit the regulaturs would near to the side wall of the case.
but i also heard, that its best for shunt regulators to be as close as possible to the dac-board to keep the cable from shunt to dac as short as possible.
if i am going to cool the parts with the side wall of the case, the cable would be at least 13-15cm long. too long?
(i do this only as a hobby, no technical background at all, so sorry for my perhaps very basic question.)
thanks for ideas, best horst
Hello All,
I finished a positive 1.3. I'm planning on using 2 boards to power an ACP+ preamp that has each channel on a separate PCB. Haven't completed the second board. Figured I'd try this PS with one channel.
Used a 3R resistor for R1. I connected it to a 30V transformer I had handy. LEDs lit up output voltage started to climb up to about 20V and then dropped to about 5v. LEDs dimmed and are now off. Adjusting the trimmer makes no difference. I measured across R1 and I get 0.33V.
Obviously, I made a mistake somewhere. Any thoughts on where to start? Thanks.
I finished a positive 1.3. I'm planning on using 2 boards to power an ACP+ preamp that has each channel on a separate PCB. Haven't completed the second board. Figured I'd try this PS with one channel.
Used a 3R resistor for R1. I connected it to a 30V transformer I had handy. LEDs lit up output voltage started to climb up to about 20V and then dropped to about 5v. LEDs dimmed and are now off. Adjusting the trimmer makes no difference. I measured across R1 and I get 0.33V.
Obviously, I made a mistake somewhere. Any thoughts on where to start? Thanks.
The CC limit for 0.33V/3Ω is only 110mA so the regulator most possibly current limited and dropped out. Is the across R1 legs voltage same with and without the reg's output being loaded? Normal R1 voltage drop is 0.55V-0.6V. What is the current draw spec of ACP+ per channel?
Salas, thanks for the speedy response!
The CC for each board is 160mA. Yes, across the legs of R1 it is the same, 0.33v loaded or unloaded. Loaded is with the preamp pcb. I don't have the right resistor for a dummy load.
Last edited:
In that case either Q1 or J1 or M1 have a problem. VdropR1 is essentially Q1's Vbe imposed across R1. When normally biased from J1 at 1-2mA that Vbe should be 0.55-0.6V. M1's Vgs (between pins 1,3) should be few Volt. First measure mV across R4. mV/270R should be answering 1-2mA or there about. If very little replace J1. If normal replace Q1. You need to set CC 100mA more than ACP+'s per channel current draw. Vdrop R1 should come up to normal first and use proper R1 value for that. Possibly 2.2Ω for 260mA. If its not Q1's or J1's fault, then its M1's.
Replaced R1 with 2.2R. Checked for cold joints and reflowed them all. Pulled J1 and checked it on my parts tester. Compared it to a spare and it's good. Hmm.
Put it all back together and added the preamp PCB as a load. LEDs lit but stayed Dim. 0.443v across R1. 8V instead of 5v at output. M1 quickly got hot and shut things off.
Still no mA across R4. Checked it with two different MM. Bad resistor?
Getting better but not quite there yet.
Put it all back together and added the preamp PCB as a load. LEDs lit but stayed Dim. 0.443v across R1. 8V instead of 5v at output. M1 quickly got hot and shut things off.
Still no mA across R4. Checked it with two different MM. Bad resistor?
Getting better but not quite there yet.
It's the smaller value R1 gave more power so the voltage got limited less. Maybe there's something wrong with the ACP+? Tries to go overcurrernt but the reg's limiter saves it?
See with three 1W 470Ω resistors in parallel as dummy load for 20V target output. Does the reg work correctly with this dummy load?
Is the rectified input voltage to the limiter too low or too high?
See with three 1W 470Ω resistors in parallel as dummy load for 20V target output. Does the reg work correctly with this dummy load?
Is the rectified input voltage to the limiter too low or too high?
Not if it measures 270Ω. Could be J1 doesn't pull yet due to voltage limit or because Q1 is bad.
I do not have 470R, 1 watt resistors. I put 2 270R, 3W in parallel for a dummy load. Seemed as close as I could get to the parallel value.
29V input voltage the Ubib.
Same 8V output and no mA on R4.
I'll pull Q1 and check it in the morning.
Thanks for all the support!
29V input voltage the Ubib.
Same 8V output and no mA on R4.
I'll pull Q1 and check it in the morning.
Thanks for all the support!
If Q1 proves good, next check M1. A static discharge can silently kill MOSFETs during handling or soldering.
thanks salas!
how thick would you recommend?
thanks again, best horst
edit: one last question: do you twist all cables inside a psu? wires to transformer, and from transformer to salas board and then the output to dac etc...?
Yep. Q1 is good and M1 is dead. I have the FQP3P20, so replacing it is no problem. Anything else to check?
I'll use the dummy load for startup, correct? (Maybe what I should have done in the first place.)
Yes, standard board.
Replaced M1- FQP3P20 with a working one. Set the pot to halfway. Dummy load attached at output. Checked all solder joints again. 29V AC at input.
Still same values, about 8v at output and no mA at R4. LEDs are lit but faint. I'm turning it off quickly after gathering those measurements. I'm not leaving it cook.
The flaw must be somewhere else up the line. I'll pull the Q's and J's that I haven't checked and try them.