EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: krby on April 27, 2024, 04:47:08 pm
-
I am very much an amateur electronics person, with a bit of G.A.S. I'm posting here to get opinions on my approach to buying another PSU.
I've got a 6+ years old Eventek 30V/10A PSU. I've used it for a variety of things, from powering RPis, Arduinos, other small circuits to testing strings of 12 or 24V LEDs, a stand in for 2S to 4S LiPo battery, a crude charger for a variety of battery chemistries, whatever. I don't often need beyond 3 or 5A, but have sometimes. I generally haven't been using it in places where the ripple (IIRC, somewhere around 200mV) mattered or I wasn't sophisticated enough to know it mattered.
Recently, I've been thinking about and doing more projects that I think will benefit from a better bench supply. My thinking is I would keep my existing PSU for anything that really needed >5A because I can't imagine a project I would do in the next few years that needed >5A and needed the accuracy and low noise. I'm also had a few I wanted 48VDC and just used a fixed swithing PSU (meanwell) I already have for another project, but would be nice to have this on the bench supply.
So, all this leads me to thinking about not-expensibe linear supplies that can do about 60V and 5A. This could be a one channel device, or multi channel 30V/5A that could do parallel/series across channels. I get a multichannel w/paralleling PSU, I guess I could ditch the current Eventek because that would be 30V/10A with cleaner output.
Since I'm not willing to spend a ton on this, it has me looking at the "best of the cheap" or "bottom of the quality" end of things. The usual suspects. Keep in mind I've been ok with 200mV ripple now. When I have need accurate voltage, I have verified it with a known good DMM and adjusted the supply until my DMM hit the number I wanted. So maybe the new PSU having voltage sense and feedback would be good?
- Siglent/Rigol The DP832 being the common one I see recommended here
- Korad KA3005 or KA3305 for multi output.
- Maybe super cheap RD6012P (it looks like in the 0-6A range it acts just like a 6006P) ? Should I even bother with these sort of "20mV ripple SMPS" things or just go linear?
I'm not sure I'd actually use the multiple channels all that much. But maybe that's just because I don't have them now.
Is there anything else I should be considering? I don't know if I need memory slots, I'd use them for some common configs I guess, but not required. Since I'm upgrading I would like to be able to have some better input than just coarse/fine pots. But I guess that's not required if it saves a ton of money. I haven't looked deeply at what the computer interfaces do with these things, generally I'd say I would do without that. I haven't come up with complex control needs, mostly "set to a given max current, get the voltage I want, go". Then maybe "vary current and voltage a bit manually" if I need to see how things behave.
Thanks for any advice.
-
If you have the space you could pick up something like an Agilent 6624A for very little money on ebay.
This particular unit has four isolated channels and they were used in automated test equipment setups.
They do not have banana plug sockets and the UI isn't great (LCD is hard to read) as they were meant to go into test rigs or system use and were usually remote controlled.
They are of high build quality, but are obsolete, so many installations have taken them out of service once Agilent (Keysight) stopped servicing them. All of this makes them a great deal.
If you don't feel like having to repair it from the get go, you should probably not get the absolute cheapest. Look for good vendor ratings.
-
If you have the space you could pick up something like an Agilent 6624A for very little money on ebay.
This particular unit has four isolated channels and they were used in automated test equipment setups.
They do not have banana plug sockets and the UI isn't great (LCD is hard to read) as they were meant to go into test rigs or system use and were usually remote controlled.
They are of high build quality, but are obsolete, so many installations have taken them out of service once Agilent (Keysight) stopped servicing them. All of this makes them a great deal.
If you don't feel like having to repair it from the get go, you should probably not get the absolute cheapest. Look for good vendor ratings.
Thanks for the reply! What would the advantage be to something like the 6624A vs a new (but cheap) linear supply? I guess better accuracy, regulation and ripple than a Korad? Since I'm new at this (and likely don't know what I'd be missing) I thought 1mV-3mV ripple is great for what I have planned. I really would like to be able to control it easily without a laptop or PC attached.
-
In my opinion, for a bench supply I could not see the point of getting a 6624A or similar, even if you had GAS. It's big, heavy and loud. You don't get any of the modern features like USB, ethernet and web browser control. These negatives also apply to many of the other HP/Agilent supplies that frequently get mentioned.
https://www.eevblog.com/forum/testgear/agilent-6624a/ (https://www.eevblog.com/forum/testgear/agilent-6624a/)
The Rigol DP800 & now DP900 series units are popular but the circular keypad might take some getting used to. These are also not necessarily "cheap", but maybe one will hit the clearance rack or eBay. Both can be hacked:
https://www.eevblog.com/forum/testgear/need-help-hacking-dp832-for-multicolour-option/ (https://www.eevblog.com/forum/testgear/need-help-hacking-dp832-for-multicolour-option/)
https://www.eevblog.com/forum/testgear/rigol-dp900-dp2000-series-hack/ (https://www.eevblog.com/forum/testgear/rigol-dp900-dp2000-series-hack/)
If you want to flex your budget a bit, check out the GW Instek GPP-4323. It has some oddities as well but otherwise could be of interest. It also has a load function. It seems there are various permutations with jack & ethernet options.
https://www.eevblog.com/forum/testgear/just-bought-a-gw-instek-gpp-4323-power-supply-first-impressions-etc/ (https://www.eevblog.com/forum/testgear/just-bought-a-gw-instek-gpp-4323-power-supply-first-impressions-etc/)
https://www.eevblog.com/forum/testgear/gw-instek-gpp-4323-422471/ (https://www.eevblog.com/forum/testgear/gw-instek-gpp-4323-422471/)
-
Ok, so what I'm inferring is that rather than looking at "better end of cheap" like the Korads, I should be looking at the cheap end of better brands like Rigol, Siglent, and others.
Also, any advice on 30V/5A x1 or stretching to get a multi-output parallel/serial-able that I could then treat as a full replacement of my 10A cheapo PSU?
-
Ok, so what I'm inferring is that rather than looking at "better end of cheap" like the Korads, I should be looking at the cheap end of better brands like Rigol, Siglent, and others.
Also, any advice on 30V/5A x1 or stretching to get a multi-output parallel/serial-able that I could then treat as a full replacement of my 10A cheapo PSU?
Lowest cost OK is SPD3303C but prefer the bit more expensive SPD3303X/X-E models.
Outputs can be paralled or seriesed for 30V @6A or 60V @3A. All outputs are isolated. They can also be configured as a tracking + supply.
The cheaper single output, SPD1305X and SPD1168X are 30V x 5A and 16V x 8A.
Multiple units can be placed in series for output as high as 500V max above mains earth, eg. 2x SPD3303 can provide 120V @3A. < Done this and its accepted practice endorsed by Siglent.
-
DP832. Most used gen purpose power supply on the bench (out of a lot of power supplies).
-
If I needed a new bench PSU, I'd choose between the GPP-4323 and the hackable SPD3303X-E. Budget would most likely be the deciding factor.