EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: TurboTom on December 28, 2017, 01:39:14 pm
-
Recently I've been searching for an inexpensive, moderately high voltage laboratory power supply. Since once in a while I work with vacuum tubes or discharge lamps or similar toys, and the ripple / noise of the average electrophoresis power supplies is really sub-par and moreover, these devices are quite picky of the load they are connected to (the latter can usually be worked around...), I needed a more "lab-style" instrument.
After searching the web for quite a while, I stumbled over the small PSUs made by "Bejing Hanshengpuyuan Technology Co. LTD". These devices look really simplistic as if they originate from the 1980s but the specs and description impressed me. They are supposed to be switchers with a linear second stage regulator. So I decided to have a go with the "HSPY-400-01" which is a 400V 1Amp unit, available for the equivalent of 200USD, give or take a little.
Two weeks later I received the parcel, unfortunately had to pay impot VAT & customs fees but what the heck, it still was cheaper than anything locally available in this class, provided it worked as advertised.
Initially I was really amazed by the size of this unit, it's almost unbelievable that it should be possible to squeeze 400W out of a box that small in decent quality. It was also a surprise to me that the whole enclosure is made up as a welded steel sheetmetal construction, there's not the usual plastic front/base with a folded sheet metal cover. This may contribute a little to it's dated appearance but it should provide pretty good RFI shielding and be quite robust mechanically.
Of course, I took it apart before I turned it on! Man, this thing is full of electronics and quite a complex beast. Maintenance-wise it is not far from a nightmare I guess but so far there didn't arise the need for this. The device contains a total of three switchers (one for the control circuitry, one for the fan -- yes, the fan got an individual SMPS which also provides power to the isolated RS232 control port --, and of course the main power switcher). The device appears to be equipped with a PFC circuit, and as mentioned in the description, there is a second stage linear regulator -- impressive! Moreover, there's EMI reduction stuff everywhere, maybe a total of six common-mode chokes distributed over the PCBs (and one hanging in "free air"). I found the insulation clearance around the fan SMPS PCB to the metal enclosure to be almost non-existant so I applied some thick kapton tape in this area -- should be okay now.
After reassembly, the functional test reveals that the power supply performs really well. The voltage is spot-on and the current calibration looks also quite good as well. Set and read voltage / current are displayed all continuously on a single two-line 16 chracter LCD. Power isn't displayed unformtunately. The command set of the RS232 interface isn't documented (at lest not in any format that would be explanatory to me, the A4 sheet of paper that I would interpret as a manual in chinese writing refers to "Modbus" at one place). I loaded the PSU with 0.9amps at 250V setting to check the ripple. I found it to be round about 70mVss at the switching frequency of 110kHz (frequency is load-dependant). There is still some superimposed common mode noise in the ballpark of 0.4Vpp that may be related to anxious me running the thing off an isolation tranformer and not having the case grounded. If I touched the PSU's enclosure and with the other hand the oscilloscope's ground, this noise was reduced considerably. I did a high-pot test of 1kV between AC input and earth and didn't observe any leakage (> 2GOhm). The PSU comes supplied with a set of alligator clip leads as well as an RS232 cable and an IEC power cord with an unidentifyable plug at the other end. An adapter is incuded that interfaces this plug to a (continental) european wall outlet, but this lacks PE so I just threw the mains cord and the adapter in the bin and replaced them with a proper cord.
So far I'ld say the power supply is worth the money though it may be wise to ckeck the isolation clearances as I did. I'll report any additional findings that may come up during use. Attached some photos of the teardown.
Cheers,
Thomas
Edit: Typos...
-
Thanks for the review - I am interested in such a supply too. It looks like they are also available on the Amazon UK marketplace - ships from China and is ~10% more expensive, but as it is sold on the UK site in GBP I think this means that the seller deals with VAT/customs fees? (Would need to check this to be sure!).
The "unidentifiable plug" is probably the standard Chinese 10A plug with pins in a Y type shape. This is similar to the NZ/Australian plugs (can normally be interchanged), but replacing it with a non-chinese IEC lead is probably wise!
-
Nice find, the only downside I can think of is the fan, looks like one of thoose high RPM ones that produce a high pitch whine, but other than that, looks good.
-
The fan is actually temperature controlled by an NTC bolted to the heat sink of the linear regulator's pass transistor. It's barely audible at idle or low power operation.
Yes the mains connector is exactly as you described, Hydron. Maybe I should keep the cable in case I'll ever get in the "parts of the woods" where this plug fits... ;).
Actually it appears that I was at fault with the PFC I mentioned in my introductory post. The inductor that I mistakenly identified as the PFC boost coil actually is part of the tank circuit of the (as it seems) half-bridge LLC topology converter. It's quite funny that the manufacturer utilized this topology despite the large output voltage range of the PSU.
Cheers,
Thomas
Edit: Added new information
-
Another thanks for the review. This HSPY company sells an amazing range of power supplies with different ratings - just search aliexpress.com for "HSPY". We bought about a dozen lower voltage supplies for our university's second year laboratory - which were really great for a year or two until some official noticed that they did not have a C-tick or CE type approval sticker on them! This means that they cannot be used in the uni and so we had to "dispose" of them all (which was very beneficial to the electronics hobbyists among the staff!).
It seems you have purchased one quite recently - so I wonder if you can check whether they have started putting CE marking on them yet. I would guess that they would need this in order to sell them on Amazon UK as Hydron mentioned. I wonder whether you have any advice to offer with regard to safety and CE type approvals?
One thing we did notice with the lower voltage units that we used was that they had a pretty large capacitor across the output - which means that if you set them to current limit at say 20mA, you can get a much higher current than what you set when the capacitance discharges. For instance you might want to push 20ma through an LED - so provided you set it up for that and turn it on with the LED already connected all is fine. But supposing you jiggle the connection to the LED so that it momentarily becomes disconnected and the voltage rises rapidly (as a current source is supposed to), then when the connection is made again the LED flashes brightly and dies!
-
Same situation with the HSPY-400-01: There's considerable capacitance wired directly parallel to the output - a total of 70µF with an ESR of roughly 1 Ohms (measured). At 400V this equals 5.6 Joules of energy, enough to destroy quite something and produce some frightening sparks... If powering sensitive equipemnt that requires a current source, adding a series resistor may be a good idea.
There is neither a CE nor a CCC marking on the device and I consider the isolation clearances around the fan power supply marginal. It won't pose any substantial danger, though, since the casing is completely made of metal and properly connected to earth. If something goes wrong, it will just blow the fuse and/or trip the mains circuit breaker and maybe destroy itself... This lack of clearance can be easily corrected with some insulating tape (kapton) or sheet (nomex/polyester). Otherwise, the design appears to be sound.
-
Dear All,
I bay Hspy 0-400V 0-2.5A
I have this product and are interested in giving me some Windows software in English?
Seler have only of chinese langue :-(
You would be very grateful
Thank you very much
-
Recently acquired a HSPY-400-01 which unfortunately came without any documentation on the RS232 command set. After quite a hunt I found it on the manufacturer' s site. Attached is a google translated version of the Chinese manual. It turns out that the command set uses the ModBus protocol and as such the RS232 interface cannot be used with a terminal emulator.
Also attached is my code (written in alps, a mix of APL and LISP) used to test the comms which works just fine and should help decode the documentation.
cheers,
-Dave
-
Just a quick report - I did end up grabbing one of these, my conclusions using it are basically identical (i.e. quite positive) to the OP's other than the fan in mine is a bit louder than I'd like at idle.
It also turns out that they have a couple of extra modes: "W" and "AH". The former allows max power to be set/displayed in addition to max voltage/current, and the latter I assume is for battery charging (have not tested it).
I did find what appears to be a bug when you change from W mode back to normal ("N") mode with power as the current selected changeable value - you can't do anything without going back to W mode, selecting voltage or current, then finally selecting N mode again.
I grabbed mine off Amazon UK for 165 GBP (third party seller) and it arrived in 2 weeks from China, though I may have gotten lucky on shipping time. They can be had a little cheaper from Aliexpress but it'd be a tossup whether you'd have to pay VAT to get it into Europe (not an issue buying VAT-inclusive off Amazon) - if they pinged you then the VAT + admin fee + hassle would cost more than the difference in price.
-
It also turns out that they have a couple of extra modes: "W" and "AH".
Thanks for mentioning that. I haven't noticed that in the perfect Chinese manual.
I've bought 3 versions - 30V 10A; 400V 1A and 1000V 0.3A. They are really small and portable. For not so demanding tasks I prefer these than my pro bench PSUs that weight a ton.
I've attached some footage of 30V version and 1000V version (400V version looks the same as in the first post so no point showing it here).
The controller is based on STM8S005CBT6. There is also CS1180S 24bit ADC made of pure chinesium. Converter is in fact double stage - first isolated SMPS and then linear stage. I love how they did some pieces of the circuit on separate PCBs and attached them vertically - it is quite tightly packed.
1000V version has a bit questionable isolation but it works (until it fails and electrocutes someone). Fortunately, the case is made of thick metal and properly earthed so it should be reasonably safe.
Overall quality is surprisingly good. The PSUs are a bit slow to reach set voltage (sometimes 1-2 seconds) but accuracy is very good. There are some calibration options in the menu and it is also possible to set small voltage and current offset. Power limitation in "power mode" is slow and not very accurate.
-
I am trying to control this Power supply via RS232, I read the manual, seems MODBUS allright. but tried all iterations of Speed, 7/8 bit, stop bit = 1/2, big endian, small endian.
but to no avail, the power supply just sits there and being a snob. |O
I recon I can sniff the control protocol if there is anybody who can share the PC software that could control the PS via RS232?
-
Hi,
a colleage at my company obviously bought a hspy-300-03 via alibaba.
I don´t know the price but a short search showed offers around 450$.
Anyway he returns in a couiple of weeks and so did I pick it up.
When I read the printing on the cartonage I immediately knew I had a real quality device in my hands :palm:
On opening the Minus connector fell out of the box .... cleanly broken off .... great quality struff seemingly pressed from mud. |O
First impression ... well, ahem yes
Weird IEC-connector and even weirder adaptor without PE!
No CE sticker ... not even the fake chinese CE.
Second impression .... the casing is not folded to perpendicular shapes, but the front is slightly tilted in the horizontal ... and it doesn´t appear to me as if its related to the broken connector.
Anyway .... I opened it up to see the innards.
Large PCB on the second floor bends easily as there are several isolation slots across ... flimsy thing
Some inductances flying around .... mechanical construction leaves alot to desire.
Lets inspect the PE .... certainly not as it must be done ... just screwed to the IEC-connector fastening screw wo. washer and fixation et al. The lid of the casing also just relys on that 5 mashine screws will somehow make it through the laquer. :-//
Small to almost non-existant isolation distances all over the place. |O
When my eyes fell on the big brown caps I first was positively surprised to see the well known NipponChermicon/UnitedChemicon logo ... well, until a closer inspection revealed that it were a case of blatant copyism .... Iron KingKong .... I almost fell from my chair LMAO :-DD :-DD :-DD
I admit I never heard of that ... KingKong caps I ´m still giggling. :-DD
KingKong even at the output ..... and of course bigger in value than healthy ::)
The wiring also doesn´t look as if it could cope with teh specced 900W continuosly.
At that point I decided just to put that piece of crap back into its carton, place it aside and write a warning to my colleague rather to throw that thing into the garbage directly or at least to not switch it on, before taking it apart.
So far I'ld say the power supply is worth the money though it may be wise to ckeck the isolation clearances as I did.
No, I disagree fully.
A device which is inherently unsafe, which is a possibly lethal accident waiting to happen, is certainly not worth any money at all.
Just keep off of that crap ... and best return the garbage back to it´s sender.
I´m beginning to think that dumb ol´ Trump might be right with his taxation plans against china
Taxes and customs can´t be high enough to prevent such crap beeing sold to decent countries.
Not to think of what useful could have been built from those now wasted resources. :wtf:
regards
Calvin
KingKong ... I still can´t get it :-DD
[attach=1]
[attach=2]
-
Hello,
I'm bringing this topic back up from the grave, as I came upon those supplies (especially those with voltage higher than 100v), I read every comments here, it seems that most like them, besides one person.
I don't really trust such a supply with an output exceeding 400w, it's already a lot I think. The 400v 1a one is the one I want.
Besides that cap, which can be replaced maybe, and some insulation problems (maybe it can be fixed too, with some tape as someone did ?) it looks quite appealing on the paper.
Is there anyone else who own it ? For this price, it's really hard to beat. The minimum I want is at least 200v, and 1amp. I can't find any offering where the price isn't skyrocketting. I need a decent stability, as I want to use it (not only for that) as a source to calibrate an electronic load, and I want my supplies to be stable at least to the 0.02% in specs. Is the noise high ? Are specs in general close to real life ?
Thanks guys ! 8)
Lemon
-
HSPY is a good power, made in Beijing, not the small factory of southern China.
-
Can you elaborate a bit please ? :) I don't know much about the different quality between various places in Asian countries. The only new chinese pieces I have currently is an XJW01 LCR Meter, which I quite like for the price (I wonder if I can adjust it with better resistors as references to improve accuracy), an Uniroi 30v/5a power supply (quite noisy, not really a fan, it was my first psu) and I bought recently a kp184 load, which seems to be quite nice, especially for the price.
-
Don't mean to derail this thread, but are any of the 400 V power supplies available with US 120 V input power? I can't seem to find any.
-
I looked around for you and I didn't find any either :/.
I wonder how stable those 400v supplies are. The load regulation spec states 0.02%, which is decent, if that's true.
-
I looked around for you and I didn't find any either :/.
Aw rats. Thanks for looking as well. Now the search continues. Asking myself if I want to get a step up transformer to use the HSPY, or just find another bench supply in that higher voltage range.
-
Hello everyone! I boght a 1000V 1A version of this power supply. This is really usefull, good and cheap device. But after a day of experimenting i broke it, short circuit on the hot part (see photo). As a result of a short circuit, the output diodes of MUR460 burned out, and two HV mosphets 2SK3748 too, on the "hot" part burned two mosphets 47N60C3 and one gate drive mosphet [or bipolar] transistor I unsoldered it (the inscription can't be read, see photo of backplate). So output diodes MUR460 are too weak, two diodes are connected in series (one diode is designed for 600v, a two consecutive: for 1200v), for impulse load this is not enough and i think this caused the failure. So i think full repair cost is about 100$ :-//
-
Don't mean to derail this thread, but are any of the 400 V power supplies available with US 120 V input power? I can't seem to find any.
I just ordered 110vac versions of the 1000v and 400v 1A supplies from here: https://www.aliexpress.com/item/1005001288362565.html (https://www.aliexpress.com/item/1005001288362565.html)
I added a note to ship 110vac versions, and double-checked with the shop to confirm. This is the response, which apparently is automatically translated
"Because usually this power supply is 220V, 110V we make extra, a few people will buy 110. We make it by ourselves in the factory store. Please rest assured that our quality will be sold at the lowest price."
I guess that means that they buy a few 110vac versions of these. Or maybe it means they modify them? I guess I'll find out. The price is a bit higher than the 220v version.
I received both power supplies. I don't see any aftermarket mods, they appear to be factory built and designed specifically for 120 vac input. Grounding is OK, but I removed some paint under screws and ground straps to be sure they are safely grounded. Digital voltage setting is spot on. The 1 kv supply has Nichicon and Rubycon electrolytics in it. Attached is windows software supplied by the vendor. It runs under wine, allows selection of the com port and power supply model (it covers a couple dozens different HSPY models). The software works, but is rather clunky. If anyone has a way to control it from a linux script or c program I would be grateful to receive it.
-
Hello everyone! I boght a 1000V 1A version of this power supply. This is really usefull, good and cheap device. But after a day of experimenting i broke it, short circuit on the hot part (see photo). As a result of a short circuit, the output diodes of MUR460 burned out, and two HV mosphets 2SK3748 too, on the "hot" part burned two mosphets 47N60C3 and one gate drive mosphet [or bipolar] transistor I unsoldered it (the inscription can't be read, see photo of backplate). So output diodes MUR460 are too weak, two diodes are connected in series (one diode is designed for 600v, a two consecutive: for 1200v), for impulse load this is not enough and i think this caused the failure. So i think full repair cost is about 100$ :-//
did you short the output of the power supply? Or something inside the unit? Any decent current limited bench supply should be able to handle a short circuit. In fact, that's how you set the current on some of them.
How about replacing the MUR460's with an RURD4120 4a 1200 fast recovery diode?
-
Hello.
I bought two HSPY power supplies some years ago. One is 400V-2.5A and the other is 50V-20A. I had an "explosion" issue with the 400V model. Just adjusted 400V output voltage and loaded with 300W for some minutes. Before that, I rarely have gone more that 300V-350V output voltage. When I took it apart to see what had happened, I sadly discovered that the power supply employs 400V electrolytic capacitors at the linear regulator output and at the switching regulator output. :palm: The output capacitor well in parallel with the output jacks simply exploded. It had survived until now since I rarely surpassed the 350V before.
Luckily, I have identified almost all the other damaged parts: output transistors IRFP460 and TSP60R190S1, discharge current ancillary diode UF4007, discharge current regulator opamp LM358 and a small 1K 0805 SMD resistor. I do not yet discovered the value of the blown TH film resistor, marked in the attached pictures, used to set the discharge current of the TSP60R190S1 MOSFET. It looks like a 100ohm 1/2W 1% resistor, but I am not sure. My version on the HSPY 400V-2.5A seems to have a PCB layout and some components a bit different from the ones shown in previous pictures. The 300V-3A model shown previously in this thread seems to use a 10ohm 2W resistor for that function, while my 50V-20A model uses a 5.1ohm 2W resistor.
Perhaps you have a similar model and can help me to find out that resistor value
Many thank!
-
I have the 400V/1A version, and the resistor is indeed 100 ohms. There are certainly differences between the two models, though the main one appears that the board in yours is longer to fit another SMPS on the end near the fan.
I should probably change the 400V cap in mine, unfortunately I hate working on it as it is very hard to get apart - I have already modded it by:
- removing Y-capacitors between mains (AC) and the output (increasing the ones between earth and the output to compensate) to massively lower mains leakage via the capacitors
- blocking the upper row of vents to force the airflow past the hot bits instead of bypassing them (the top heatsink on the 1A model doesn't get warm, it's only the lower ones) - the 2.5A version seems to have the fan in a better position.
- moving the temperature sense location to one of the lower heatsinks (the one that isn't live at the switching frequency, though it still is at mains voltage so care needs to be taken!)
Edit: just saw the query about software for the unit - I wrote a driver for this software by HKJ for my 400V/1A unit, should be included in the latest version:
https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/ (https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/)
Will need some changes for other HSPY models, but the driver is very easy to change (will likely just need the limits and maybe scaling to be changed).
-
Hi Hydron, thanks for your answer.
I am surely replacing all the 400V caps by 500V units and I will also check the Y-capacitors. Another thing I would want to ask you is if your 400V-1A version uses the same TSP60R190S1 MOSFET. I noticed that it will be quite hard for me to get this exact part number and wonder if maybe your version is using some other equivalent MOSFET that could be easier to obtain. In a few days I will continue my searching for more damaged parts, but I hope to find no more. Then I will power up the power supply after having all damaged parts removed to inspect if there is still another unseen defect.
Interesting your HSPY driver for the TestController software. After repairing my unit, I will try to adapt your driver to mine.
-
Mine has a different MOSFET. I can't remember the exact connection of the MOSFET, but if it is only being used for discharging when the output is turned off then it probably doesn't matter what one is used as long as the voltage rating is OK. See attached pic.
-
Hi All
Just wondering if anyone has successfully used this power supply to grid charge a hybrid Toyota battery?
Thank you in advance
-
Hi All, i recently bought the HSPY-1000-001. Came with a nice chinese manual. Only this forum, had a tread on this power supply, and dpenkler posted some documents that helped me getting the device programmed. It took me way to many time, so i would like to share how i got it working.
You will find the info in the attached documents, the powerdc document is the dos batch file
Kind regards, hoping to help some future buyers!
-
Anybody else get tonnes of common mode noise on the output of these?
I picked up one of these a few months ago (the 1000V version) for a project I'm working on and until now hadn't even looked at the output noise as it was not at all critical for that. Last night though I tried to use it for something else and finally saw the output. I was blaming my circuit for a while until I started isolating things and realised that it was coming from the supply itself.
Basically as long as the mains plug is in, and even if the unit is off, there's a tonne of 50Hz present at the outputs. I've since idiot checked myself by plugging into every other power supply I have on hand and everything is nice and clean, but with the HSPY it's hum-town.
-
Yeah there are some Y-caps in dumb places, see an earlier post I made. If you get rid of them like I did then it doesn't hurt to put a film cap to earth from the output to fix some of the HF noise you're left with.
-
Yeah there are some Y-caps in dumb places, see an earlier post I made. If you get rid of them like I did then it doesn't hurt to put a film cap to earth from the output to fix some of the HF noise you're left with.
Are there any schematics floating around or did you just trace things out? I just popped the lid off mine thinking I might give it a tune up in a spare 5 minutes before work, but it's so densely packed around the output terminals I can't see much, and pulling it apart looks like a pain in the ass so I'm gonna go back to it later (hopefully without having to gut it and trace out too much)
-
No schematics, just looked for the Y-caps across isolation barriers (they are pretty obvious). As you've found out it's a massive pain to work on though!
If you're lucky you might be able to just do a snip-snip job on the caps without pulling it entirely to bits - it's been a while and I can't recall how badly they are buried though sorry.
-
OK, finally made the effort to pull it apart enough to get those caps out. Much better, but still not great. I don't think I could be bothered trying to improve this thing any further given the lack of schematics/pain it is to disassemble, time to build something else for the task at hand I reckon.
-
I measured the voltage between GND and the minus with the main switched turned on to 69VAC
A 10uF 400V between GND and minus pole reduces the voltage from 69VAC down to 69mVAC.
-
Deleting the existing Y-caps which were across the isolation barrier, and adding 100nF of earth - negative capacitor reduced mine to something like 400mVpk-pk. Could have put something larger in there to reduce it further, but ~100nF seems fairly standard and what's left seems low enough to ignore.
-
comes now with a CE sign...