Service Manual: eevblog.com/fil... What's inside the Manson SPS9400 40A 3V to 15V Switchmode Power Supply? Also known as the Jaycar Powertech MP3090 and BK Precision 1692
I have been using one of these in the automotive business for 10 years now as a power supply to clip across the car battery when working on the car with the ignition on. It will run the electrics and charge the battery all day long. It never gets warm,charges a battery in a few hours,you can see the current fall from 20 to 30 amps to zero so you know how much it has charged.It will supply a genuine 40A,and is great value,branded Rapid Electronics in UK about £120. It will not start a car with a flat battery ,but leave it charging for 10 minutes then you can. Great review of a great power supply !
13.8V is also useful for charging 12V lead-acid batteries. 14.4V charges the battery faster, but you need to watch it and disconnect when it's full, but with 13.8V you can leave the charger connected all the time (if you also connect a load to the battery you just made a UPS).
Linear is better, unless you need the high current in small size and weight provided by a switchmode. Linear have less noise, and are simpler and easier to fix down the track.
@EEVblog, Dave, In Europe PFC is mandatory (according to EU standard EN61000-3-2) so it probably wasn't just some "voluntary" improvement, but something they had to add in order to sell worldwide. They did gone to town by adopting Active PFC, but most likely the inductor would be way too bulky for passive PFC to be feasible, given the power output of this unit..
It should be doing 40A in whole voltage range. The output current is eventually limited by low voltage filtering inductors. This it the best quality SWPS I've ever seen. Excellent engineering and manufacturing, except for the case! they even included filtering and snubers all over the circuit. . I am building my 1kw 24V 40A SWPS recently(also plans to use TL494, this chip is very useful, not only in SWPS, but its transistor output can only give 200ma current max . If you look for driving mosfet directly, checkout SG3525A). The manson schematic could be a very good reference for me.
7:30. I have come to realise that whenever you see a transformer before the bulk caps, it is 99% PFC, especially if there is a diode and a driver/swutch next to it.
I know this is a 7 year old video, but I have owned one of these exact supplies (branded manson), for 8 years. It got a recapping 1 year ago when I keyed up my IC7300 and the supply went bang. This was just a capacitor exploding. The supply shut down and didn't hurt my radio at all. Recapping this thing isn't trivial, as to remove the board you also have to remove the transformer and choke, making the board a bit unweildy to work on. I replaced all the capxcon caps with a much higher quality brand, the name of which escapes me but chemicon is in the name. Gave it a good clean up (7 years of dust build up) and a new, quieter fan. It still goes to this day, and I use it for battery charging on the occasion the sun doesn't want to play ball with my solar panels. I just wish it would do 30V for charging my 24V LFP pack!
That is a mean PSU and really well built, I could do with something like that at work, I have to make do with Manson PSU's but 0 to 30V at 2.5 Amps. They do well but recently I've needed 28 Volts at 15 Amps and have had to put 4 of these in parallel being very careful to match the voltages before switch on. If they do a version which reaches 30V that would be something I would buy. Nice tear down Dave.
See my comment below. Basically these are paired inside the same case and produced in the same process, thermal runaway of one of them is not really an issue due to good thermal coupling and very similar tolerances.
I would like to see more videos like this, for us young players (yeah! 51 years old). Great that Dave goes through showing the schematic and the board, explaining as he goes. More videos like this!
Yes and no. It's generally not a good idea to parallel diodes because of thermal runaway of one diode due to production tolerances. But the word "thermal" together with a look at the datasheet should give you a hint. These diodes are stacked so both of them (D15A and B) are in the same case. This in turn means that firstly their tolerances are very similar (since they are produced in the same process) and secondly the thermal coupling between them is very good.
C41 which is in series with the main power converter transformer is to prevent flux walking of the core. If the push-pull switching has even a small offset, then flux walking may result which will end with core saturation and popped switching devices.
The cool thing about the power supply is it uses standard components we can get anywhere. The schematic is very typical, and very similiar to how I design it too, and its easy to fix!
I've got a Manson EP-850, 13.8V DC 50A linear. It's the size of a domestic hi-fi amp and weighs about the same, bloody thing's a monster! You wouldn't be throwing it around on screen like the 9400 :D
I am in the process of repairing one of these (badged Lowe SPS8400), contstruction quality is adequate but not great, thanks for the the link to the PDF. Mine has been sitting around kaput for years, about time I fixed it.
Very impressed it has power factor correction too, which is required overseas, they could have gone all the way and used sync rectification, get that efficiency well in the high 90's...still very good.
The circuit of the meters isn't shown in the schematics, not even the type is given. But there are trimmers on the back of the meter board. Look around there. I guess some of the resistors will either form a voltage or set an amplification within the meters. Maybe by exchanging (switchable) some of them change the scale (decimal point will be in the wrong place then though). But don't turn the trimmers if you haven't got something to recalibrate the meters! (not that easy at 40 amps)
I 2nd this, I got about 8 of em in my closet from various systems I did over the past 10 years, I'll either recycle em some day, or try to fix them for older systems. I was always a little nervous to open them up though as I have read the caps in them can be deadly.
Me too. TV power supplies and monitors. They don't have any decent lifespan. 1/3 of my bad caps in my removed cap collection are CapXon. I'm not sure what makes CapXon particularly bad, but whatever it is, they are completely inept at producing a decent capacitor it seems.
Ah! I have a slightly newer version of this power supply, a Tenma 72-7665. They have completely removed the power factor correction portion of the circuitry! :O They've also whittled it down to 2 heatsinks instead of 3 (due to the removal of the power factor correction mosfets Q1 and Q2), relocated the metal can input filter to the rear panel, and probably a few other minor changes that have escaped a cursory glance. I also only paid $125 for it on ebay. :D
Ah yes, you are correct. I mistook the capacitors in front of them for heatshrink on a two lead diode. I guess co-packaged diodes are well matched enough that they can be used this way?
I was with you on the high quality of this supply until near the end I noticed that the main output cap is a CapXon. I've probably replaced hundreds of those on PC motherboards...
The schematic shows two diodes but I only see one on the PCB, and certainly only one high current one for each side (on the heatsink). Not sure why it's like that but it doesn't look like they're actually paralleling diodes.
Thats actually sps9400 as 8400 version is with analog meters. But over $300 holy crap, that same thing sells with maas branding in finland for 150 euros including vat.
I used 2 server power supplies to warm up a room and to power everything in that room, too. I used it until a month before this comment got posted because one of the power supplies decided it wanted to be faulty, and because of that, I almost got deleted from existence. So dont service faulty power supplies. Server or not. They have enough power running thru them that they will delete you in under half a second.
Hey @EEVblog +EEVblog Dave - Any chance you could do a video on dangers of working around high-voltage caps, how to safely discharge say 240VDC caps (using a 60W filament bulb?) and other safety tips please? Thanks!
be nice if there was an inspection panel on the underside of the unit ,make it easier to replace faulty parts its a bit messy having to take the pcb of the main chassis to do any work on the unit
I was curious how the auxiliary power supply from transformer TR2 worked since it seemed to be fed with DC from C10/C11. So that little 3-pin device TOP222Y is switcher too - never knew such a device existed - thanks.
Cool! Looks like a nice unit. I could only see two wires coming out of the main transformer... According to the schematic it's a centre-tapped secondary so there should be three wires. Maybe that's the difference between the two model numbers? Or maybe I just couldn't see the third wire? Interesting vid!
thanks a lot! Can you pay more attention to PFC module? It's pretty interesting to understang how does it works, and it's a pity that no one explains it in youtube!
I wonder what would happen if only you accidentally short-circuit alligator-probes. Big bang... But I am interested on whether putting two hi-current diodes (D15A+D15B, D16A+D15B) in parallel without additional series resistor for each of diodes is a good practise? I mean resistors for current-balancing or current-sharing.
Thank you that makes sense. As a follow up... Why use two diodes? i.e. Why not just have one larger diode in that one package? I'm assuming that as the physical size of the junction gets larger the overall forward rating would increase. So in this case two diodes allow for more current but why isolate them into distinct diodes in that package only to use them in parallel? Hopefully my question makes sense.
Hi all. I am fascinated by SMPS's but am still learning. Can you guys tell me how the feedback circuit on this PS works and how the PWM duty cycle can be adusted if there is too much output voltage. Is this what the manual is describing earlier on. Is there a feedback circuit or is it user adjusted. I don't get this bit. Here I am looking for optocouplers or feedback transformers. Bear with me - please explain. Brill vid once again.
Hi, sorry to disturb you, can you help my power Maas SPS 9400 40, the cooling fan always runs, immediately when it is turned on it runs and never stops, what should I do to change the thermostat, where it is located, on the component diagram. Thank you
Hi Dave I have the 25A version of this one. Do u know if there is a way of moveing the desimal point on the current meter to the left 1 place so that it will work at a higher res for readings of up to 9.99 amps? I was thinking I could find a way of makeing it switchable or something?
Dave - I love your videos. Thank you for taking time to make them. Keep in mind I'm very much a novice with this stuff so take it easy on me... But why is it ok to have the parallel diodes (D15A, 20DL2C; D16A; 20DL2C) after the TR4 transformer? Doesn't one diode end up taking all of the load?
dave can you make a video about how accurate this psu is? also what is that pieces of equipment next to ur bench psues? a signal generator maybe? (the black one with the screen and the big knob)
The Power Mosfet IRFP460 is rated 20A@25°C or 13A@100°C continuous drain current and 80A pulsed drain current. Which current value should one consider with a switching PSU; the continuous or the pulsed one?
V Good explanation. How clean is the output? If you had shown on scope that would be really worth noting. Which smps? is very Stable having clean output like linear power supply under heay currents upto 5 amps or more?
I have BK Precision 1692 but my PCB and schematic are slightly different. My supply won't come on. No fan, display or anything. I am not getting a VCC out of TR1. I'm having a difficult time identifying/ordering a replacement transformer for TR1. Any advice? Thanks, Mike
Does anyone have a 120V/60 Hz version of this device and can take a picture of the PCB for me? I modified my Peaktech 1540 (another OEM for the Manson SPS9400) 240V/50 Hz supply quick and dirty by replacing the bridge rectifier with a voltage doubler. I am curios what route the manufacturer went in their 120V/60Hz derivative.
I have been using one of these in the automotive business for 10 years now as a power supply to clip across the car battery when working on the car with the ignition on. It will run the electrics and charge the battery all day long. It never gets warm,charges a battery in a few hours,you can see the current fall from 20 to 30 amps to zero so you know how much it has charged.It will supply a genuine 40A,and is great value,branded Rapid Electronics in UK about £120. It will not start a car with a flat battery ,but leave it charging for 10 minutes then you can. Great review of a great power supply !
Aw you never plugged it in. I wanted to see if it could cope with 40A @ 15V continuous using the "power supply killer".
13.8V is also useful for charging 12V lead-acid batteries. 14.4V charges the battery faster, but you need to watch it and disconnect when it's full, but with 13.8V you can leave the charger connected all the time (if you also connect a load to the battery you just made a UPS).
Linear is better, unless you need the high current in small size and weight provided by a switchmode. Linear have less noise, and are simpler and easier to fix down the track.
@EEVblog, Dave, In Europe PFC is mandatory (according to EU standard EN61000-3-2) so it probably wasn't just some "voluntary" improvement, but something they had to add in order to sell worldwide. They did gone to town by adopting Active PFC, but most likely the inductor would be way too bulky for passive PFC to be feasible, given the power output of this unit..
It should be doing 40A in whole voltage range. The output current is eventually limited by low voltage filtering inductors.
This it the best quality SWPS I've ever seen. Excellent engineering and manufacturing, except for the case! they even included filtering and snubers all over the circuit. .
I am building my 1kw 24V 40A SWPS recently(also plans to use TL494, this chip is very useful, not only in SWPS, but its transistor output can only give 200ma current max . If you look for driving mosfet directly, checkout SG3525A). The manson schematic could be a very good reference for me.
7:30. I have come to realise that whenever you see a transformer before the bulk caps, it is 99% PFC, especially if there is a diode and a driver/swutch next to it.
I know this is a 7 year old video, but I have owned one of these exact supplies (branded manson), for 8 years. It got a recapping 1 year ago when I keyed up my IC7300 and the supply went bang. This was just a capacitor exploding. The supply shut down and didn't hurt my radio at all. Recapping this thing isn't trivial, as to remove the board you also have to remove the transformer and choke, making the board a bit unweildy to work on. I replaced all the capxcon caps with a much higher quality brand, the name of which escapes me but chemicon is in the name. Gave it a good clean up (7 years of dust build up) and a new, quieter fan. It still goes to this day, and I use it for battery charging on the occasion the sun doesn't want to play ball with my solar panels. I just wish it would do 30V for charging my 24V LFP pack!
Thumbs up if you would like Dave to talk about and use one different unit of his test gear every friday!
That is a mean PSU and really well built, I could do with something like that at work, I have to make do with Manson PSU's but 0 to 30V at 2.5 Amps. They do well but recently I've needed 28 Volts at 15 Amps and have had to put 4 of these in parallel being very careful to match the voltages before switch on.
If they do a version which reaches 30V that would be something I would buy.
Nice tear down Dave.
Tuesdays are crap at work... having this to watch at home makes the day a bit better :)
See my comment below. Basically these are paired inside the same case and produced in the same process, thermal runaway of one of them is not really an issue due to good thermal coupling and very similar tolerances.
i have had my Manson SPS9400 for years and its always worked well.
Great power supply and really robust. Used one for 7 years when testing Icom amateur radios with zero probs.
I would like to see more videos like this, for us young players (yeah! 51 years old). Great that Dave goes through showing the schematic and the board, explaining as he goes. More videos like this!
Yes and no. It's generally not a good idea to parallel diodes because of thermal runaway of one diode due to production tolerances. But the word "thermal" together with a look at the datasheet should give you a hint. These diodes are stacked so both of them (D15A and B) are in the same case. This in turn means that firstly their tolerances are very similar (since they are produced in the same process) and secondly the thermal coupling between them is very good.
Loving tear down Tuesday! Been waiting for a consistent video from you and this delivers in spades!!! Keep up the great work Dave!
C41 which is in series with the main power converter transformer is to prevent flux walking of the core. If the push-pull switching has even a small offset, then flux walking may result which will end with core saturation and popped switching devices.
The cool thing about the power supply is it uses standard components we can get anywhere. The schematic is very typical, and very similiar to how I design it too, and its easy to fix!
Im lovin' Teardown Tuesday! Great job Dave!!!
Normally, yes, but these are two devices in the same package, so would be fairly well matched.
I've got a Manson EP-850, 13.8V DC 50A linear. It's the size of a domestic hi-fi amp and weighs about the same, bloody thing's a monster! You wouldn't be throwing it around on screen like the 9400 :D
I am in the process of repairing one of these (badged Lowe SPS8400), contstruction quality is adequate but not great, thanks for the the link to the PDF. Mine has been sitting around kaput for years, about time I fixed it.
Very impressed it has power factor correction too, which is required overseas, they could have gone all the way and used sync rectification, get that efficiency well in the high 90's...still very good.
The circuit of the meters isn't shown in the schematics, not even the type is given.
But there are trimmers on the back of the meter board. Look around there. I guess some of the resistors will either form a voltage or set an amplification within the meters. Maybe by exchanging (switchable) some of them change the scale (decimal point will be in the wrong place then though). But don't turn the trimmers if you haven't got something to recalibrate the meters! (not that easy at 40 amps)
I forgot to mention they are in the same device (only two devices on the heatsink, not four), so would be fairly well matched.
I 2nd this, I got about 8 of em in my closet from various systems I did over the past 10 years, I'll either recycle em some day, or try to fix them for older systems. I was always a little nervous to open them up though as I have read the caps in them can be deadly.
Me too. TV power supplies and monitors. They don't have any decent lifespan. 1/3 of my bad caps in my removed cap collection are CapXon. I'm not sure what makes CapXon particularly bad, but whatever it is, they are completely inept at producing a decent capacitor it seems.
It takes time for CZcams to process the HD version.
Ah! I have a slightly newer version of this power supply, a Tenma 72-7665. They have completely removed the power factor correction portion of the circuitry! :O They've also whittled it down to 2 heatsinks instead of 3 (due to the removal of the power factor correction mosfets Q1 and Q2), relocated the metal can input filter to the rear panel, and probably a few other minor changes that have escaped a cursory glance. I also only paid $125 for it on ebay. :D
there is a OHP sheet(thin transparent plastic sheet) over the Paper
Ah yes, you are correct. I mistook the capacitors in front of them for heatshrink on a two lead diode. I guess co-packaged diodes are well matched enough that they can be used this way?
Ah, thanks for sharing, didn't know that, hadn't heard of CapXon. I'm surprised, because everything else seems decently done.
HD video works here if you click the gear icon to change the quality.
I was with you on the high quality of this supply until near the end I noticed that the main output cap is a CapXon. I've probably replaced hundreds of those on PC motherboards...
The schematic shows two diodes but I only see one on the PCB, and certainly only one high current one for each side (on the heatsink). Not sure why it's like that but it doesn't look like they're actually paralleling diodes.
Thanks for the tear down Dave!
I've been waiting for this one :). Thanks for finally getting around to it.
You seem so much younger and more inquisitive in these early videos.
I was looking at getting one of these from Rapid electronics, so interesting to see what is inside.
Have one on the desk. This one is a Maas SPS-9250, only 25A, but not too bad a price around 120 GBP.
Thats actually sps9400 as 8400 version is with analog meters. But over $300 holy crap, that same thing sells with maas branding in finland for 150 euros including vat.
I used to use high temp lacquer on televisions when a customer complained about coil noise.
Don't know. you'd have to look at the panel meter circuit in detail.
About TL494, it's not a common application schematic, it's a test circuit. Of course, it doesn't shows us much about application.
I used 2 server power supplies to warm up a room and to power everything in that room, too. I used it until a month before this comment got posted because one of the power supplies decided it wanted to be faulty, and because of that, I almost got deleted from existence. So dont service faulty power supplies. Server or not. They have enough power running thru them that they will delete you in under half a second.
@ fourbarposer, a blob of hot melt glue or epoxy perhaps?
Coils only of course. As you know It'll do bugger all to help crap caps.
Nice shield screens for the tranformers! Briliant!
really good video, loved how in-depth you got with it!
Hey @EEVblog +EEVblog Dave - Any chance you could do a video on dangers of working around high-voltage caps, how to safely discharge say 240VDC caps (using a 60W filament bulb?) and other safety tips please? Thanks!
It would be as "accurate" as you calibrate it. I'm sure the temp-co of the reference would be reasonable enough for the job.
THAT clear stuff is not heat shrunk. It is clear vinyl standard tubing.
NICE the way there is pink glue on all the mounting screws.
that TL 494 is also used in several PWM Flyback drivers :3
be nice if there was an inspection panel on the underside of the unit ,make it easier to replace faulty parts
its a bit messy having to take the pcb of the main chassis to do any work on the unit
I was curious how the auxiliary power supply from transformer TR2 worked since it seemed to be fed with DC from C10/C11. So that little 3-pin device TOP222Y is switcher too - never knew such a device existed - thanks.
Ah, no sorry, it's three wires, the third is a different colour. I forgot to mention that.
Hi
Can you do teardown of the SPS-9602, how it can be improved and adding the current output control?
Regards
2002 ? Wow they were ahead of their time with the active PFC etc...
Those packages on the heatsink actually house two diodes each.
My Lowe SPS 8400 has just died, this video is interesting, Thank you
Cool! Looks like a nice unit. I could only see two wires coming out of the main transformer... According to the schematic it's a centre-tapped secondary so there should be three wires. Maybe that's the difference between the two model numbers? Or maybe I just couldn't see the third wire? Interesting vid!
No, but you could charge the battery pretty quickly.
thanks a lot! Can you pay more attention to PFC module? It's pretty interesting to understang how does it works, and it's a pity that no one explains it in youtube!
There are quite a shitload of rebranded ones out there like BK Precision etc... Easily found a few on element14 ... Great video nonetheless !
Excellent job, Nice PSU also !
I wonder what would happen if only you accidentally short-circuit alligator-probes. Big bang...
But I am interested on whether putting two hi-current diodes (D15A+D15B, D16A+D15B) in parallel without additional series resistor for each of diodes is a good practise? I mean resistors for current-balancing or current-sharing.
Its perfect for running car amplifiers on the bench
We call the copper strip around the transformer a "belly band"...
Thank you that makes sense. As a follow up... Why use two diodes? i.e. Why not just have one larger diode in that one package? I'm assuming that as the physical size of the junction gets larger the overall forward rating would increase. So in this case two diodes allow for more current but why isolate them into distinct diodes in that package only to use them in parallel? Hopefully my question makes sense.
Hi all. I am fascinated by SMPS's but am still learning. Can you guys tell me how the feedback circuit on this PS works and how the PWM duty cycle can be adusted if there is too much output voltage. Is this what the manual is describing earlier on. Is there a feedback circuit or is it user adjusted. I don't get this bit. Here I am looking for optocouplers or feedback transformers. Bear with me - please explain. Brill vid once again.
Yeah, I agree. It's not the best look!
oh forgot the description was excellent
I hope to be well filtered for HAM use because HF equipment is sensitive! I am a ham and know it from practice!
Hi, sorry to disturb you, can you help my power Maas SPS 9400 40, the cooling fan always runs, immediately when it is turned on it runs and never stops, what should I do to change the thermostat, where it is located, on the component diagram. Thank you
The planet spins pretty quickly, it can make people a bit dizzy and unsure of what day it actually is. Beware.
Hi Dave I have the 25A version of this one. Do u know if there is a way of moveing the desimal point on the current meter to the left 1 place so that it will work at a higher res for readings of up to 9.99 amps? I was thinking I could find a way of makeing it switchable or something?
Hey, Dave!
What power supply is better for hobby, switchmode or linear?
Thank you very much for this interesting video
Excelent information and video, thanks and congrats.
I like to see one that is at least 40 amps and 40 volts. that what I like to see were I can control both the voltage and current.
I was waiting for this :) Thanks!!!
Can someone explain how he said "Warning Will Robinson" in 2012 when netlflix show "Lost in Space" came out in 2018 O.o
Dave - I love your videos. Thank you for taking time to make them.
Keep in mind I'm very much a novice with this stuff so take it easy on me... But why is it ok to have the parallel diodes (D15A, 20DL2C; D16A; 20DL2C) after the TR4 transformer? Doesn't one diode end up taking all of the load?
dave can you make a video about how accurate this psu is?
also what is that pieces of equipment next to ur bench psues? a signal generator maybe? (the black one with the screen and the big knob)
Ugh! CapXon on the output! Those caps are rubbish - I can't count how many I've replaced.
bob's my dad....
HI, just viewed your excellent video but you incorrectly stated the three resistor pre-sets as capacitors.(25:55)
I had 2 of these, both krapped there self in 1 years of use.
Yeah, I cant discuss on eevblog forum, my account isnt approved by two months...
Dave what about you do a teardown on your Constant Current Generator
I have to say it : you're good.
I noticed some gobs of hot glue on the power switch. Is it absolutely taboo to have hot glue across both terminals?
The Power Mosfet IRFP460 is rated 20A@25°C or 13A@100°C continuous drain current and 80A pulsed drain current. Which current value should one consider with a switching PSU; the continuous or the pulsed one?
@EEVblog, Dave, am I seeing a bad cap C8 in the PFC circuit? 7:16 Looks like a stain in the pressure relief groves.
Could you modify the Power Supply to be current limiting ? If so, how would you do it?
V Good explanation.
How clean is the output?
If you had shown on scope that would be really worth noting.
Which smps? is very
Stable having clean output like linear power supply under heay currents upto 5 amps or more?
I have BK Precision 1692 but my PCB and schematic are slightly different. My supply won't come on. No fan, display or anything. I am not getting a VCC out of TR1. I'm having a difficult time identifying/ordering a replacement transformer for TR1. Any advice?
Thanks,
Mike
Does anyone have a 120V/60 Hz version of this device and can take a picture of the PCB for me? I modified my Peaktech 1540 (another OEM for the Manson SPS9400) 240V/50 Hz supply quick and dirty by replacing the bridge rectifier with a voltage doubler. I am curios what route the manufacturer went in their 120V/60Hz derivative.
I couldn't find the NTC reference. is a power surge protector ?
Good stuff!
That front panel makes it look like a cheap piece of stereo equipment from the 90s.