Hi
Mills is very good still some builders maintained that they could pick Caddock MP930 as best choice there.
You can use thinner wiring when on 4 wire configuration because its there to null the cables resistance. The sense wires can be even thinner than the force pair.
You can jumper the sense pair at the connector, that will render the connection 2 wire conventional. Its up to you to judge what difference does it make in your application.
Normally 4 wire config has a thing for resolution going for it on the subjective level.
You are welcome.
Mills is very good still some builders maintained that they could pick Caddock MP930 as best choice there.
You can use thinner wiring when on 4 wire configuration because its there to null the cables resistance. The sense wires can be even thinner than the force pair.
You can jumper the sense pair at the connector, that will render the connection 2 wire conventional. Its up to you to judge what difference does it make in your application.
Normally 4 wire config has a thing for resolution going for it on the subjective level.
You are welcome.
Thanks to Salas and TeaBag for the design and GB respectively - much appreciated.
I am doing a lot of reading through this thread getting my head around my first salas build. I have a few 10000uF 40V LCR caps kicking about and wondered whether they would be OK to use for C105/205/305 position? They look like this:
Just wondered whether anyone had come across these and what they thought...or am I better off getting a mundorf mlytic?
Many thanks,
Crom
I am doing a lot of reading through this thread getting my head around my first salas build. I have a few 10000uF 40V LCR caps kicking about and wondered whether they would be OK to use for C105/205/305 position? They look like this:
Just wondered whether anyone had come across these and what they thought...or am I better off getting a mundorf mlytic?
Many thanks,
Crom
Last edited:
That one in the example pic is 1988 vintage. Very good quality when new but they can be well expired by now. Check the YY-WW label print on yours. You need an LCR that measures dissipation factor to be sure about their state. Anything normal ESR spec with >0.25 D and anything LOW ESR spec with >0.1 D goes to the dustbin.
You would also need snap-in type for ease of build in this PCB.
You would also need snap-in type for ease of build in this PCB.
Hmmm, thanks for the information Salas. The date on mine is October 1996...so a little bit younger!
I had to do some research to understand what you meant here. In case it's useful for others, I found this useful:
Sci.Electronics.Repair FAQ: Capacitor Testing, Safe Discharging and Other Related Information
Sci.Electronics.Repair FAQ: Capacitor Testing, Safe Discharging and Other Related Information
I don't have an LCR but it appears that I can use this procedure to check whether I have some duds or not: Sci.Electronics.Repair FAQ: Capacitor Testing, Safe Discharging and Other Related Information
I can't find specs for these. Am I right in saying that these should match a normal ESR spec?
I had to do some research to understand what you meant here. In case it's useful for others, I found this useful:
Sci.Electronics.Repair FAQ: Capacitor Testing, Safe Discharging and Other Related Information
Sci.Electronics.Repair FAQ: Capacitor Testing, Safe Discharging and Other Related Information
I don't have an LCR but it appears that I can use this procedure to check whether I have some duds or not: Sci.Electronics.Repair FAQ: Capacitor Testing, Safe Discharging and Other Related Information
I can't find specs for these. Am I right in saying that these should match a normal ESR spec?
Yes, those are normal ESR spec capacitors. A really good brand, I still have two 15000uF/63V cans of that line in one of my amps. A JLH MOSFET class A/B.
Good useful info links. Such reservoirs are way more complex than beer cans...
Electrolytics should be measured at 100/120HZ for DF and uF value by the way.
Good useful info links. Such reservoirs are way more complex than beer cans...
Electrolytics should be measured at 100/120HZ for DF and uF value by the way.
Thanks Salas. Following a lot of reading I am well on my way to a weekend full of D factor & IDss measuring.
I was hoping someone wouldn't mind giving me a 'BOM check' as this is my first one. I've made some calculations and 'borrowed' a few ideas from others in this thread who are also using it for my application.
I am aiming to build 2 +/- pairs to run a Legato 3.1 i/v stage (individual L/R supplies with jumpers J1-J4 removed - no power to centre SE stage). This means +/- 15v with a constant current of 350mA and I'd like around 150mA headroom in the shunt.
I've made the following calcs:
Transformer: 20-24vac @ 80VA
R101 Current setting resistor: 5.6 ohm, 2W (or higher)
R103: 1.8k ohm
R105 1k trimmer
C102 Mundorf MKP at 10 uF
C104 Zobel cap: Mundorf MKP 4.7uf
4 x green LEDs
2 x red LEDs
Load resistor: 42r for 350mA, 150r for 100mA (around 5W)
Inevitably, I have a few additional questions:
1) The current setting resistor: I've calculated the value to be 5.6r but I read that a Caddock Mp930 was the resistor of choice here. Parts Connexion are pretty expensive for these but Mouser has a 5R 30W in stock. This will push the constant current to 620mA. Alternatively they have 7.5R which means current sits around 450mA. My inclination is that I'd rather have a little more headroom above the required 350mA here. Does anyone who uses these with Legatos have any comments on which would be a more appropriate (assuming that I'm able to dissipate the heat - 5.6W - if I opt for 620mA). But then again 100mA may be more than enough!
2) I noticed that schottky diodes came in for a bit of stick earlier in the thread. Is this just purely because they are schottky or is there a specific incompatibility with this circuit that I should consider? I can't think what but I'm up for learning! I have a few Cree Silicon Carbides that I ended up not using in another project that I thought I could roll into this one.
Thank you in advance for any corrections and feedback!
Crom
I was hoping someone wouldn't mind giving me a 'BOM check' as this is my first one. I've made some calculations and 'borrowed' a few ideas from others in this thread who are also using it for my application.
I am aiming to build 2 +/- pairs to run a Legato 3.1 i/v stage (individual L/R supplies with jumpers J1-J4 removed - no power to centre SE stage). This means +/- 15v with a constant current of 350mA and I'd like around 150mA headroom in the shunt.
I've made the following calcs:
Transformer: 20-24vac @ 80VA
R101 Current setting resistor: 5.6 ohm, 2W (or higher)
R103: 1.8k ohm
R105 1k trimmer
C102 Mundorf MKP at 10 uF
C104 Zobel cap: Mundorf MKP 4.7uf
4 x green LEDs
2 x red LEDs
Load resistor: 42r for 350mA, 150r for 100mA (around 5W)
Inevitably, I have a few additional questions:
1) The current setting resistor: I've calculated the value to be 5.6r but I read that a Caddock Mp930 was the resistor of choice here. Parts Connexion are pretty expensive for these but Mouser has a 5R 30W in stock. This will push the constant current to 620mA. Alternatively they have 7.5R which means current sits around 450mA. My inclination is that I'd rather have a little more headroom above the required 350mA here. Does anyone who uses these with Legatos have any comments on which would be a more appropriate (assuming that I'm able to dissipate the heat - 5.6W - if I opt for 620mA). But then again 100mA may be more than enough!
2) I noticed that schottky diodes came in for a bit of stick earlier in the thread. Is this just purely because they are schottky or is there a specific incompatibility with this circuit that I should consider? I can't think what but I'm up for learning! I have a few Cree Silicon Carbides that I ended up not using in another project that I thought I could roll into this one.
Thank you in advance for any corrections and feedback!
Crom
Last edited:
1. Calculations are always ballpark because of LEDS Vf and MOSFETS VGS tolerances. You just have to apply what you calculated and then verify the current by dividing the voltage drop across the setting resistor via its ohmic value. From there on you can correct with proper value if the CCS current proves grossly out of target causing heat problems. If its not heating a given sink too much when on the plus side, leave it there, its good for the output impedance. So start with a plain vanilla resistor and order a hi spec part only after you verify the proper value.
2.Schottky diodes are no problem, to the contrary they are clean and low loss, as long as they exceed the rectified DC RMS on the big filter capacitors by times four at least in their peak inverse voltage spec. They are normally not as strong in that aspect as common diodes are but for low voltage stuff there are many that will suffice.
2.Schottky diodes are no problem, to the contrary they are clean and low loss, as long as they exceed the rectified DC RMS on the big filter capacitors by times four at least in their peak inverse voltage spec. They are normally not as strong in that aspect as common diodes are but for low voltage stuff there are many that will suffice.
Hi Salas,
My farther update:
1. I added a choke 2H 70 Ohm 10mA in between Raw DC and my MC (LT317). In details: it is at "+" side in between 4700uF cap located Raw DC PCB and entrance to of LT317 with 1000uF electrostatics. Hiss is 90% gone. I have mild white noise in max volume now. That is it.
2. Dies it make sense to add 4 1000pF snubber caps to rectifier diodes in Raw DC (MUR120) to min impulses?
I'm looking for Wima MKP10 1000pF 5% (0.001uF) 1000V.
Thank you.
My farther update:
1. I added a choke 2H 70 Ohm 10mA in between Raw DC and my MC (LT317). In details: it is at "+" side in between 4700uF cap located Raw DC PCB and entrance to of LT317 with 1000uF electrostatics. Hiss is 90% gone. I have mild white noise in max volume now. That is it.
2. Dies it make sense to add 4 1000pF snubber caps to rectifier diodes in Raw DC (MUR120) to min impulses?
I'm looking for Wima MKP10 1000pF 5% (0.001uF) 1000V.
Thank you.
1. This is where I added my choke and I changed C1 to 1000uF 50v Elna RFS.
2. Understood.
Attachments
Last edited:
The rectifyer spikes can be killed/lowered with caps/snubbers to ofc but chokes are cool in the way that rightly chosed and dimensioned they just cant pass this interference. They are not capable. And choke input also protects the backfeed to the transformer and out to other devices. If one has the voltage, its a fairly easy solution to cleaner PSU. Imho
No, this is what printed on my choke. Do think I need a bigger one, lat say 50mA?