Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

ArcticMyst Security by Avery | Browser Hide by Avery

Small batch of 4A buck drivers coming soon in early October.

Giannis_TDM

Giannis_TDM

Well-known member
Joined
Apr 27, 2019
Messages
927
Points
93
Hello there! Due to some demand I keep getting I will have about 15 drivers spare to sell here, Current is pre set by me and each driver is individually hand assembled and tested by me, Specs are:
Max current out 4.5A
Recommended max current out 4A
Vin 3-16v
Vout = Vin - 0.2v
Reverse input polarity protected
Size: 15.75mm x 9mm, All Components purchased off mouser, caps are Samsung X5R or better dielectric. Inductor is Panasonic automotive grade line.
Price per driver 20 euros/25USD.


Any questions or 'pre orders' can be sent below, thanks for looking 🍻
image.png
image.png
 





Giannis_TDM said:
Hello there! Due to some demand I keep getting I will have about 15 drivers spare to sell here, Current is pre set by me and each driver is individually hand assembled and tested by me, Specs are:
Max current out 4.5A
Recommended max current out 4A
Vin 3-16v
Vout = Vin - 0.2v
Reverse input polarity protected
Size: 15.75mm x 9mm, All Components purchased off mouser, caps are Samsung X5R or better dielectric. Inductor is Panasonic automotive grade line.
Price per driver 20 euros/25USD.


Any questions or 'pre orders' can be sent below, thanks for looking 🍻
image.png
image.png
Click to expand...
I mean X7R would be so much better than X5R here considering just how hot it runs. Not even saying that this is temperature rating, not hte most important parameter for switching mode converter.

What's the efficiency, ripple and start-up curves?
 
atomd said:
I mean X7R would be so much better than X5R here considering just how hot it runs. Not even saying that this is temperature rating, not hte most important parameter for switching mode converter.

What's the efficiency, ripple and start-up curves?
Click to expand...
Thanks for your feedback! As you can imagine not having the drivers as per I do not have the boards yet it is quite a challenging task to provide what you are asking for, from the previous revision: typical efficiency in low 90s, up to 94% at 14vin and 1.7A out, current ripple sub 2mA at 2A, voltage ripple 16mv at those amps and the curves you can find on the datasheet of the open source buck on my website, the architecture is the same just updated for more current.

I do agree X7R would be better, Sadly no company does actually make 25v 22uF 0805 caps with a X7R dielectric, Best is X6S, the caps that are available in X7R are X7R, Also PS forgot to add,

You are mistaken X7R and X5R are not only the temperature ratings, It is a different dielectric. Class 2 ceramic caps like these have a voltage dependent capacitance, i.e their capacitance can change with the applied input voltage, When the input voltage has reached the cap's voltage rating the capacitance would have been nearly halved, X7R also specifies that the aforementioned effect is minimized compared to X5R. This is absolutely one of the most important things to consider for a switch-mode converter, As you can imagine for a given inductor of x and a switching freq of y having the output capacitance suddenly halve could cause large issues as far as ripple is concerned. Besides that as you mentioned, For X7R the capacitance change per degree C is less so when putting both effects together the change can be quite drastic. Hence why the output capacitance is split into different value cap banks in parallel so it can also enjoy the better response of having bulk capacitance and lesser ones to improve ESR, ESL and freq response in regards to filtering. hope this helps!
 
Last edited:
Giannis_TDM said:
Thanks for your feedback! As you can imagine not having the drivers as per I do not have the boards yet it is quite a challenging task to provide what you are asking for, from the previous revision: typical efficiency in low 90s, up to 94% at 14vin and 1.7A out, current ripple sub 2mA at 2A, voltage ripple 16mv at those amps and the curves you can find on the datasheet of the open source buck on my website, the architecture is the same just updated for more current.

I do agree X7R would be better, Sadly no company does actually make 25v 22uF 0805 caps with a X7R dielectric, Best is X6S, the caps that are available in X7R are X7R, Also PS forgot to add,

You are mistaken X7R and X5R are not only the temperature ratings, It is a different dielectric. Class 2 ceramic caps like these have a voltage dependent capacitance, i.e their capacitance can change with the applied input voltage, When the input voltage has reached the cap's voltage rating the capacitance would have been nearly halved, X7R also specifies that the aforementioned effect is minimized compared to X5R. This is absolutely one of the most important things to consider for a switch-mode converter, As you can imagine for a given inductor of x and a switching freq of y having the output capacitance suddenly halve could cause large issues as far as ripple is concerned. Besides that as you mentioned, For X7R the capacitance change per degree C is less so when putting both effects together the change can be quite drastic. Hence why the output capacitance is split into different value cap banks in parallel so it can also enjoy the better response of having bulk capacitance and lesser ones to improve ESR, ESL and freq response in regards to filtering. hope this helps!
Click to expand...
First of all you're right but also wrong. Yes X7R and X5R are obviously different dielectric material but the name specifies ONLY thermal properties. As a matter of fact if you open mouser and search for murata X7R caps is same package and same capacitance and look them up in murata's capacitor explorer you'll see that they can much different DC-bias characteristics (which also explains why 0805, 16V, 22uF X5R can differ in price ten times).

I'm not sure what do you mean by different capacitors but as far as I can see on your webpage you're using single capacitor package. Mixing different values of capacitors in parallel in the same package is TERRIBLE idea. Not only you get smaller filtering capacitance then if you'd just use highest value available but if you throw it into any field solver you'll see nasty parallel resonances that greatly reduce filtering at certain frequencies. Sometimes it matters, sometimes doesn't but paralleling different valued capacitors is practice from old, THT times and currently only applicable in deep RF with field solver
 
atomd said:
First of all you're right but also wrong. Yes X7R and X5R are obviously different dielectric material but the name specifies ONLY thermal properties. As a matter of fact if you open mouser and search for murata X7R caps is same package and same capacitance and look them up in murata's capacitor explorer you'll see that they can much different DC-bias characteristics (which also explains why 0805, 16V, 22uF X5R can differ in price ten times).

I'm not sure what do you mean by different capacitors but as far as I can see on your webpage you're using single capacitor package. Mixing different values of capacitors in parallel in the same package is TERRIBLE idea. Not only you get smaller filtering capacitance then if you'd just use highest value available but if you throw it into any field solver you'll see nasty parallel resonances that greatly reduce filtering at certain frequencies. Sometimes it matters, sometimes doesn't but paralleling different valued capacitors is practice from old, THT times and currently only applicable in deep RF with field solver
Click to expand...
Huh, Did not know the rabbit hole went that deep, The cap values on these drivers are based on purely maths and testing, with my scope I have found that 4x 22uF 25v // with 2x 2.2uF 25v caps deliver the best ripple performance and least noise across various currents for the driver. Though as it appears you are vast more knowledgeable in this subject I can see your point, I will search about field sim tools to see if there are any decently priced options to add into my development process, Thanks again for the feedback! Though that aside, These drivers are just meant to be just cheap and high quality for the community, not 0.001% precision current sources, I have made them to the criteria of less than 1% current ripple, less than 1% voltage ripple and more than 90% efficiency, since they hit these criteria with flying colors (for instance voltage ripple is usually about 0.5%, said for a typical setup of 2A out @ 4.6v) I am a happy camper.
 
Last edited:
First testing performance, @ 1.6A out @ a diode vf of roughly 5.31v, 8.4vin from my bench PSU. 6mV voltage ripple and 3.15mA current ripple over a 0.4 ohm shunt. That is 0.19% voltage ripple and 0.12% current ripple, Pretty nice!
IMG_20241007_121842.jpg
Voltage^
IMG_20241007_122821.jpg
Current^
 
Oh almost forgot, efficiency measured to be 92.47% in this case, (8.4vin, 5.7vout (higher since I used the DMM6500's shunt to measure the current at the same time in series with the load, 1.635A out, 1.2Ain), Here are some pics :)
IMG_20241007_120504.jpgIMG_20241007_120405.jpg
 
atomd said:
Why don't you just order PnP?
Click to expand...
Well for small batches I don't mind soldering them, hot air is quite fast and I've gotten decent at manual PnP, Besides that the overhead from the assembler and the faster speed are why I choose to hand assemble, along with ensuring each component is handled right and I am not getting the assembler's 5 year old IC with the humidity indicator in the pink. Just guarantees components are nice and recent from the factory line and in general I enjoy doing it. Having assembled more than 200 of these lads with no reported issues from customers means I am doing ok :)
 
You must log in or register to reply here.





Back
Top