EEVblog

Teardowns of the Rohde & Schwarz HMP4040, HMP2020, and NGE103 programmable lab power supplies.
Unboxing video: • EEVblog #1173 - Rohde ...
Comparisons with other brand PSU's: There are some comparisons with other brand PSUs:
www.rohde-schw...
www.rohde-schw...
www.rohde-schw...
Forum: www.eevblog.com...
Bitcoin Donations: 38y7DE8HEHNj8fGDtUr4PkCn9nWxiorvvy
Litecoin: ML7oQokTwB38bgzzjLDbRV97HKAHuwRfHA
Ethereum: 0x11AceA38DCA9DbFfB4F35f3F746af65F9dED28ce
EEVblog Main Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
Support the EEVblog through Patreon!
/ eevblog
AliExpress Affiliate: s.click.aliexpr...
Buy anything through that link and Dave gets a commission at no cost to you.
Stuff I recommend:
kit.com/EEVblog/
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.co...
T-Shirts: teespring.com/s...
Likecoin - Coins for Likes: likecoin.pro/@...
#RohdeSchwarz #PowerSupply #Teardown

[@peterstalin7463]

Why is this teardown not on pornhub???

[@EEVblog]

I got banned, it was too hardcore.

[@GeneralPadron]

LOLOLOLOLOLOL, oh man.....

[@basshead.]

You should make a PC themed porn called ''Motherboards Get RAMmed''.

[@0toleranz]

EEVblog uhh you gotta dispute this. Transformer metal is always soft core iron for better magnetic flux 😲

[@jagardina]

It's there. Just filter by crazy Austrian.

[@finkelmana]

I can think of two reasons they built it to that size. The first, is to keep it the same size as other equipment. Obviously, these are stacking units, so perhaps they will stack with other/previous models that can not be made smaller. Secondly, is purely cosmetic. Perhaps if it was smaller, people might think it wasnt as good as a larger model.

[@EEVblog]

Yeah, likely for stacking/physical compatibility

[@DarkFire515]

I guess the spare volume also allows for future feature growth without having to redesign the case & chassis.

[@sageii5614]

To justify the ridiculous price for the poorly engineered product.

[@cacanovotny]

Yes, that's it. I used to work for a while for R&S in Vimperk and you would be surprised how much empty space is in some of the instruments (I remember some 5.5 digit DMM that was 95% free). But the form factor is kept the same for stacking.

[@km5405]

a bit of space is not bad per se either. lots of room for air to take heat from other parts away, and lots of separation between different parts of the circuit so they don't interfere as easily.... this is not designed to be a particularly portable device....so no need to compact it all down.

[@marsgal42]

Number One Rule of Power Supply Design: if you can lift the power transformer it isn't big enough. :-)

[@michaelathens953]

Also true of guitar amp transformers

[@H-77]

Same goes for the audio transformer.

[@bobbybiggs4348]

The relay on the mains board might be soft start for inrush current. It is after all a fair sized toroid.

[@simontay4851]

Yes, that is what it is for. The relay will activate after a few seconds.

[@electronash]

Oww. I knew somebody would beat me to it. lol The NTC is likely in parallel with the relay contacts, so yep, inrush soft-start.

[@electronash]

And maybe for an overcurrent trip as well, although you'd think they would have to rely on the secondary side relays and transistors anyway for that. I guess they could be doing remote power-on as well, but that would leave the NTC roasting hot while the power switch is on.

[@Arek_R.]

Yep soft start

[@romainf145]

USB bodge wire to earth is probably because they failed CE Safety certification that requires a low impedance return path to mains earth. Quick and easy fix.

[@seeker4430]

Evertime a r&s product is opened I lose a little more respect for the brand

[@sren2385]

@@seeker4430 yeah...you have to know it

[@erhanozdogan4510]

Thanks to Dave & Rohde & Schwarz for giving new power supply design ideas...

[@spmanojgowda]

"swing a shrodinger's cat in there" 😂

[@reprapmlp]

As a designer of toroidal power transformers in a previous life, the [lack of] clearance between case and bolt in the first one gave me nightmares. Shorted turn faults are scary stuff.

[@MrOrangeman18]

I’m sorry Dave but that first one you teared down, that shorted turn issue is totally unacceptable on something of this cost. If you stacked another supply on top then it would no doubt short out. And if this was on another psu say a $200 one you would be slating the hell outa it. Unbelievable

[@EEVblog]

No, you are wrong, it does not short out if you stack another unit because the weight points are on the outside tabs. To hit the bolt you have to press directly in the center, and press fairly hard.

[@MrOrangeman18]

EEVblog yea that’s fair enough. But would you not agree that it’s a poor design/ was overlooked? If it did short out what would be the outcome? Would it damage the psu?

[@bavarianmonkey8326]

@@EEVblog the issue is that the shorting issue depends on where you put the force. Given that in a lot of cases random devices get stacked on top of each other it does not really matter where the weight points are on any given manufacturers device. They simply might have overlooked the issue.

[@redsquirrelftw]

They should have a rubber pad or something on top of the screw. Tiny price for a bit of peace of mind.

[@MrOrangeman18]

BavarianMonkey yep but he can’t say anything bad about it as he got it for “free”

[@martin_3146]

Hey Dave, how about you turn them on and test them a bit? We just got a NGE102 and the current limiting seems to be very wonky. Current limiting seems to take a few ms to kick in and it blew the ass right out of some LEDs we tested with it. They might turn out to be not as great as you imagine. Right now i moved back to the trusted old one and will test with a load and a scope before hooking up something else to it. Also wifi on a power supply with probably Windows CE on it - Never imagined you being a fan of that.

[@Drottninggatan2017]

You could put a plastic cap on top of that transformer screw. It's better to be safe than sorry.

[@samgamgee3622]

Maybe he could use one of his couch feet. :-) Seriously though, I don't know why they don't do it at the factory. Or... use a bolt that's the right length and doesn't stick out 1".

[@KingMysion]

Agreed. The quality looks superb with this one notable exception.

[@danielwulf4241]

Do we get to see a measurement of the ripple and noise of the Rohde & Schwarz power supplys?

[@dogastus]

Data sheet inserts for the various chips etc much appreciated. Thanks Dave,

[@oriole8789]

That 1 turn short circuit definitely seems like a significant oversight. Just needs a soft cap. Lacquered metal is fine, but repeated hits will break through that coating. 3.7A on the output would mean pretty insignificant current difference on the input, but the imperfect contact in a real scenario can becomes a source of mystery EMI that could create one of those impossibly annoying/difficult to track down problems for a customer. Like the microwave oven problem at the Parkes Observatory. ;) Not a deal breaker by any means but a problem that should be taken care of in a later revision. Thanks for these teardowns. Good stuff as always.

[@innopriest]

It's not lacquer, it's powder coated. I doubt it would ever short out unless someone intentionally caused it.

[@oriole8789]

@@innopriest Hm I'm not sure why my mind went to lacquer. Powder coatings are pretty tough so you might be right!

[@MichaelBerthelsen]

When you first start talking about the MELF resistor around 8:10, I first misheard it, had to do a double-take...😂

[@esven9263]

It's so odd, really high quality electronics, but the physical design elements are all kind of bizarre. The unit cost on some of those parts even in quantity is very high but then there's no vibration resistance, no consideration for the case flex, and generally low quality construction generally speaking.

[@timothygodfrey4950]

there's no vibration resistance, no consideration for the case flex,-it sits on a bench-why bother...

[@esven9263]

​@@timothygodfrey4950 It has to be shipped at some point which can be a fairly aggressive process even if it's handled carefully. Long or heavy components can experience a lot more mechanical wear from such forces than other components and capacitors in particular have a tendency to fail due to work hardening of the aluminum tabs which internally connect the foil windings to the pins. Likewise the unit may need to be moved into and out of storage, or it could be mounted on a cart and moved frequently from room to room. These aren't even theoretical usage cases, I've seen both these situations very frequently with lab power supplies. Components freely vibrating like that can also cause solder joints to crack, particularly if there is a large amount of PCB flex like there is here where the PCBs hang unsecured within the enclosure. As for case flex, he demonstrated in the video that just putting a piece of equipment on top of this one could cause the case to form a shorted single winding. Mechanical considerations do matter.

[@movax20h]

It is interesting to see switching pre-regulator plus linear main regulator getting into low noise and precission lab PSU these days. I didn't see them much in the past because of the noise, and more complex control circuit or micro controller feedback required.

[@FireballXL55]

You only got 3A because the voltage is really low, if the case touched the end of the bolt it will be a lot higher. I bet they do an amendment and have either a shorter bolt or a cap or some sort of insulation pretty quick.

[@ElmerFuddGun]

Both videos are 29:40 (or 39) long... seems odd that the tear down would be EXACTLY the same length as the unboxing. Or is there some hidden significance of 29:40 with Rohde and Schwarz or Dave? Hmmm...

[@EEVblog]

No comment...

[@c0ffeeman]

The number 24?

[@randycarter2001]

To bad R&S can't read my suggestion of putting a plastic cap on that bolt here.

[@EEVblog]

I'm sure they are reading comments. A cap would do the job nicely.

[@Monkeh616]

@@EEVblog Or just small piece of insulation sheet (the stuff used for double-insulated metal enclosures) on the shell. Either way it's a few pennies and a few seconds a unit.

[@mr_gerber]

@@hwfanatic Instead of Air, a dielectric?

[@xxJohnxx77]

I liked the NGE quite a bit more than the HMPs. Clean internal desing and no floating PCBs. Might have to get one of those one day.

[@hikaru-live]

For that bolt, they could have put a rubber cap on it that would both insulate the bolt and provide an additional mechanical support point for the case.

[@peterzingler6221]

Thanks i was disapointet there was no teardown in the other video

[@EEVblog]

It was an hour edit, so I thought better to split into two videos.

[@iamdarkyoshi]

7:02 Nice comic sans for the OptiMOS text

[@urdnal]

Blegggh. Whenever I see inappropriate use of comic sans I really would like to interrogate the person who chose to use it to ask why? Why do you think it's a good idea to use this shitty looking unprofessional joke of a font? Where's your self-respect? I still get flashbacks from the Higgs boson presentation...

[@MrKillswitch88]

Would love to see you make a video or two about mercury arc rectifiers as they are pretty cool and I like that they glow while in use.

[@soothcoder]

Do you plan to characterise these? It would be kind of cool to see how quick the voltage control loop responds to a load transient (and how much it over/undershoots) and to see how quickly after overload they go into current limit. Be interesting to see how much it differs between the units too. Hopefully they don't die like... ahem... the previous one you tested :) And it would be cool to compare wit the Rigol

[@gamerpaddy]

Lol theyre using the slow 8bit atmega64 and atmega1280 for the main business and the super beefy stm32f407 for simple networking & USB. At least theyve put a beefy ATsam 32bit mcu in the bigger model

[@Anonymouspock]

It's pretty funny. Similar to the whole ESP8266 or especially ESP32 thing where the WiFi chip is far faster than the application processor.

[@jaycee1980]

I had a serious WTF at that... an FPGA AND a grunty processor - just for ethernet, usb and wifi ? WTF ?

[@jaycee1980]

@Proscriptus that still doesnt warrant the processing power in use. You could do such things on an 8 bit micro just as easily, with fuse protection to prevent the code being read out

[@porklaser]

@@jaycee1980 That networking module probably has an RTOS and a networking stack and all that. It's also probably standardized across many product lines so it has to be flexible. Running the power supply itself is pretty simple in comparison.

[@muctop17]

Think industrial! Re-use in different product families for decades. Delivery of all components for years. Thats what they come from.

[@starlite528]

In the US, at least, 'warranty void' stickers have been proven un-enforceable in court. I wonder if they're equally as useless in Australia?

[@mr_gerber]

Interesting. Links?

[@starlite528]

@@mr_gerber www.npr.org/sections/thetwo-way/2018/04/11/601582169/warranty-void-if-removed-as-it-turns-out-feds-say-those-warnings-are-illegal

[@mbaker335]

A large box helps cooling and allows equipment to stack. Neat stacking is a big consideration.

[@MaxKoschuh]

2:25 this is a standardized format (from Hameg days)

[@JackZimmermann]

Have a feeling that if the first power supply were designed by a Chinese manufacturer, they would get a big fail for having the PCB just hanging there without any front support. And don't forget the shortage risk. I would like one of their scopes though...

[@kanetw_]

It's supported by the heatsink relatively close to the front.

[@EEVblog]

There is no shortage risk, it's fairly stiff. The risk in the flex and vibration induced issues during transit.

[@gabiold]

I would choose that design instead of the "whole motherboard" one. Easier servicing and that cooling is well thought out. This is not a mobile device, maybe 7-8cm of PCB hangs out, without any heavy component. Should be more rigid than the "big board" version which doesn't have screws at every 8cm too...

[@EEVblog]

Gábor Móczik I used to design mobile test racks that got wheeled around on a factory floor 24/7. Even with gigantic rubber wheels you'd be surprised at the vibrational modes you can get in components and boards and mounts.

[@IncreasingVoltage]

I am almost 100% sure that if these Powersupplys would have a "Jing lo yingi Electronics" name on the front, Dave would find 100 reasons to make it look s#it.

[@piconano]

You must be dreaming if you think you can buy a Chinese product made this nice, without paying the same price or more.

[@IncreasingVoltage]

@@piconano I wasn`t talking about the price. But interesting point.

[@Zetex2000]

I doubt it. It just doesn't look like cheap chiense shit.

[@diatomsaus]

To be fair, the design isn't even that great. That point which could be shorted out is damning, my illuminator is definitely heavy enough to make a good day turn bad. Also was vibration resistant considered? I can't see any. Those flappy capacitors also doesn't install much confidence, gunk them down! I expect near perfection for this price. The Rigol isn't any worse in terms of construction.

[@gabiold]

That ducted cooling should be the de facto standard. I always hated fan-cooled things, why should be dust all over inside an equipment to just cool a few parts? With some forward thinking you can design such a ducted cooling system, or mount the high dissipation items on the back, on a heatsink, and you don't even need a fan!

[@allesklarklaus147]

Only real concern to me is that not isolated screw through the transformer shorting out to the case. They gotta fix that please

[@mdbssn]

Those red wires in the NGE - they didn't have the clearance to get the trace width they wanted on the PCB around the other parts, so instead of the extra like 8mm worth of trace and gap, they just put the wire on instead.

[@starlite528]

put a strip of kapton tape on the inside of the case to keep it from shorting in worst case scenario. I love how the output transistors are mounted in a way that looks like they're easily replaceable.

[@TheDefpom]

@15:00 that glue holding down the wires looks a lot like a glue that ends up going low resistance and shorting out when it gets old and starts to break down, it’s a known issue in CB / HAM Radios which have to have that glue picked off.

[@EEVblog]

The devil is in the chemical detail. You can't tell by just looks.

[@allesklarklaus147]

The Defpom's Repair Channel The glue also looks a lot like the glue used everywhere else without any problems. Like PC power supplies for example

[@LazerLord10]

I'm not sure if the free-hanging boards are too much of an issue, as I don't think it'd be too much different than just mounting it with a few screws to a normal chassis. It's neat that the heatsink is also serving as a support!

[@EEVblog]

You'd be surprised. I used to vibration sweep test PCB's and you'd be shocked at the resonant modes of boards, mounts and components. There are commercial standards for transport vibration by various transport modes like trucks and planes.

[@pxidr]

I found the Rigol PSUs less messy in terms of design.

[@Stefan_Payne]

Why this big? Probably because of Rack Mounting capability. If you mount it in a rack, you need the lenth anyway...

[@vencibushy]

That Omron relay is probably there on the mains PCB as a part of inrush current limiting circuitry for the toroidal transformer.

[@pa4tim]

The open chassis case construction of the first one is typical Hameg (pre-R&S)

[@marklowe7431]

Reducing the volume of common components across models increases costs. It's basic commerce.

[@innopriest]

30 years apart but this is very similar to my NGPV 40/3 R&S supply.

[@heyikidebir]

What is the reason of having that big ass toroidal transformer while you already use switching regulator just before the linear regulators? Can't you use a switching regulator to generate isolated voltages to be feed into pre-regulators? I wonder the design trade off happening there. Noise and EMI performance maybe?

[@telefon72]

Fun fact, the case the first one is built in is nearly identical to some 20 ish year old hameg we have at my uni. The power button is even the same.

[@PlayerSalt]

I assume the dimensions are larger than needed just so all the products can stac , like the width needs to be the dame for racks , the depth is likely just so all the products can stack with each other.

[@piast99]

8:40 - I would check that red wire. It looks like it was poorly soldered to the PCB.

[@supercompooper]

Measure the ripple and noise! :)

[@jagardina]

They could have run all those wires from the transformer through one protective sleeve. The labor required to thread all those wires...

[@Konecny_M]

The extra labor is ZERO. The glass fiber sleeving is always used on toroids on taps going outside of the winding. Someone just got clever and specified them to be cut bit longer.

[@fluffyfloof9267]

8:37 oh hey there, is that a cold / dry joint on the red output wire? edit: 11:25 no worries, looks fine from this angle.

[@cannesahs]

What about soldering of red banana jack at 11:33 ?

[@MichaelBerthelsen]

That cooling system is pretty cool, and loving the transformer!😁👍 As for the size, maybe it's meant to pair with some existing or future system (stacking, like we've seen they've designed for), and that part explains the length? Just having a stack of power supplies stacked makes less sense than combining them with a set of other tools, at least to me. Not sure what they'd want to pair it with that requires that large a form factor, though...

[@Pops180]

They should stick to making scopes. These power supplies dont look like they are worth the money.

[@chrisb4009]

I’ve never understood the cost of power supplies.

[@EEVblog]

Try go build and sell one with this same functionality and you'll find out.

[@timh2870]

Dave, the relay bypasses the inrush limiter after a few seconds.

[@ElmerFuddGun]

_"It's going to be much more efficient and lower noise"_ - 27:04 Well you have them both so how about proving those claims in another video? I would guess it would have lower noise but efficiency is a different animal.

[@EEVblog]

A switching pre-regulator is going to beat the pants of a pure linear design in efficiency, except maybe at a small window of optimum voltage for the linear.

[@H-77]

I think you need to tear down a really big old linear PA amplifier... The transformer in this thing is nothing compared to what some of those things use.

[@radekoplustil6750]

I have one HMP4040 at home, another one at work. I guess I must have some kind of older model, because the front panel on yours is dark grey/light grey, mine are white/light grey. Good PSUs overall, the LCD is hard to read from angle though. They take up a lot of space on bench. Power terminals are somewhat flimsy. Great thing about these PSUs are isolated 4 channels which you can link however you want and this provides quite useful when, for example, testing TV LED backlight (100+ V).

[@necessaryevil8615]

Maybe they made it that large to make it easier to stack!

[@rfmerrill]

Why use a giant toroidal transformer if you're just feeding switching regulators anyway? (Honestly asking, I figure there must be a reason).

[@Konecny_M]

Many galvanically isolated windings (each output has at least two secondaries), with a bonus of free low-ish noise auxilitary rail to power the control section on each channel is why. Derived feedback from just one secondary wouldn't fly with the switcher, so you'd be effectively left with N+1 switchers which adds lots of extra noise and complexity for almost no gain in efficiency and degraded performance only to save a bit on weight on device that gets moved maybe ten times during its lifespan.

[@OneBiOzZ]

They might be doing primary side along with secondary side voltage selection switching on that first teardown (i only got that far :P )

[@pennfootball71]

kind of midfi i like point to point wiring better...take a look at Primaluna amps and integrated amps.

[@ketamu5946]

I guess, they made the lenght universal on all Units they made to have the same Side panels for one series... would make the most sense economically.

[@GeneralPadron]

There is a way to remove the warranty label without damaging it, :).

[@TheRedNickBoy]

The Relais at 4:10 could be part of a kind of soft start for the transformer, couldn't it?

[@borisp123]

8:39 Bad solder joint of the red wire?

[@dhpbear2]

9:50 - The only way to get the 'chimney effect' in that heat sink is to run the unit on its side?

[@EEVblog]

That's what the fan is for

[@3ffrige]

Half the BOM cost probably went to the Toroidal Transformer!

[@NexXxus86]

He's all drooling over that german engineering.

[@peteabc1]

The case looks exactly like every other Hameg.

[@hjups]

Are you sure that you have the placement of the OS CPU correct on the HMP series? It doesn't look like there are differential traces for the card edge, and the network card contains a CPLD, a beefy processor, and SRAM... It sounds more like the main CPU running the OS is on the "network card", and it communicates with the front panel MCU and channel control MCUs via some sort of serial protocol. As I recall, the ST ARM also has enough internal flash to store a lite linux image.

[@DrakkarCalethiel]

I saw those relays and instantly thought that They are Omron. Used them on countless projects.

[@johnyoungquist6540]

Are we trying to win the pounds/watt contest or the needless component count contest. Outrageous overkill for DC power. The mechanical and thermal design really stinks. Shake it at resonance and it won't last an hour. Lots of wires to break, boards to flex, components to fall off. Size your shunts to not require huge heat sinks. Gain is cheaper than heat sinks. Could go on and on.

[@allesklarklaus147]

JOHN YOUNGQUIST Say we spec a 0.1Ohm shunt, the power we need to get rid of at a given current is current squared times resistance, so that is (10A*10A)*0.1Ohms equals 10W. At the lower end the problem is that we have a 0.2mA resolution. 0.2mA through a 0.1Ohm resistor gives you 20nV to work with. That's a big dynamic range for a PSU right away but the real problem is that if we spec a 0.01Ohm resistor which is about what we need to get rid of the heatsink we are trying to (properly!) measure 2nV across that resistor.. At which point the heatsink will be cheaper.

[@muppetpaster]

If they made the boards any shorter you'd be complaining that they didn't leave enough space between the elements to avoid interference or somethimg. You would find something.

[@southerncharity7928]

Hi Dave, sorry for a comment on an old video. Does the HMP series overshoot in CC mode? (like the NGE103)

[@trcostan]

Has to be that length to stack with the other units :)

[@z1power]

I really prefer power supplies with quality CapXon-brand caps, none of this Nichicon rubbish that don't even leak out

[@anggorotriatmojo1200]

Heyyy 04:08.... it is OMRON... made in indonesia.... hello there fans from indonesia.... i am from indonesia.... and hello for eveyone....

[@aeroslubis5704]

finally, something from indonesia is used inside an insturement unit.

[@aicisha]

Omron isn't Japanese?

[@michelfeinstein]

Dave, even if the top doesn't fully touch the screw, won't there be some capacitive/inductive coupling?

[@pisspee2099]

What are the gold dots for? 20:03 . Also, isn't the braid on the wires to protect the insulation from heat?

[@michaelfogarty3239]

the size might be related to rack mounted systems

[@starlite528]

I wonder if the power switch is just activating the relay?

[@stuartmcconnachie]

@9:29 and @10:38 looks like they have extra pads (to the left on the output stage) also? Is this perhaps to mitigate against unavailability of the component in the required form factor? So they can use 2 of half the value in parallel if required.

[@uzaiyaro]

Wow, I’ve never, ever heard of a resistor being in a plastic IC package like that. Is this common? I’m still rather new to electronics, so I’m curious!

[@yar0nix224]

why do they have all those random gold dots? theyre not test points...

[@MatthewSuffidy]

Are those like symmetric yet isolated windings on the output of the toroidal transformer? Does that help somehow?

[@uzaiyaro]

Oh, wow. Those caps are so damn loose at 6:56. If it were me, I’d pack them with hot snot and get some rigidity into them. I did this with my DPS5030, so it’s somewhat surprising to see two power supplies at two complete opposite ends of the market have the same issue. Granted, the little Ruideng DPS units are fantastic for the money, they’re fantastic, period, but you only get so much for that money!

[@casperunnerup]

I'm confused. The screw holding in the transformer. What is that connected to and why?

[@MrTripcore]

I thought all those gold dots were test pads

[@EEVblog]

Not when there are dozens on a ground plane

[@randynovick7972]

Ah, here we go... more feeding my jealousy. Grrr. Grumble grumble. Love the new acquisitions. Good on ye. Put 'em to use for us!

[@cylejh]

Sorry to bust your balls, but those are not self tapping screws, self threading yes, but not self tapping.

[@outsideworld76]

This is the HAMEG design team at work.

[@MrFishinghook]

Why do they have so many test points on the PCB? If they even are test points.

[@Victoare]

That slurp at 17:52 :D

[@jameswyatt1304]

Try using a small X-Acto blade to cut the hole out of the warranty stickers (or feet!) so it looks better. (^_^) Also, I'm absolutely amazed at how much circuitry (huge ARM and more!) was needed on the LAN/USB board. Does it talk IEE488 on the back connector or something? I'm expecting a serial board or telnet interface, so I expected much less circuitry. Wonderful modularity and careful attention to design and manufacture. One of the most beautiful power supplies I've seen in a power supply as well. If you've got the output terminals with beefier capacitors to earth/ground, then does the one between them really matter much? Thanks for digging deeper and sharing the details!

[@telkentexas4053]

What does Dave mean when he says Bueller.... Is It something to do with Ferris Bueller's day off.

[@novafawks]

25:25 Curiousity Show, by any chance? :)

[@EEVblog]

Indeed!

[@Hexor1211]

@EEVblog What about this short connection between two points on the chassis (bolt and metal sheet) at 12:54 ? I mean, blue cover is (electrically) part of chassis, not?

[@kenarsuleyman]

When I saw transformer first time at 01:05 I was like "look at this big ass transformer" in my head and at 01:16 BOOM!

[@simontay4851]

Watch the video to the end. Now THAT is a transformer.

[@FooBar89]

I'd still use Agilent/Keysight power supplies over these, not just cheaper, but more solid engineering all around than these R&S, though competition is good and they pack quite a bit more power, would've loved to see SMPS followed by linear regulation though

[@cannesahs]

These have 50/60Hz transformer, DC/DC switching supply and finally linear regulation. DC/DC is tracking linear regulation. I'm not commenting differences between brands.

[@FooBar89]

cannesahs the AC/DC is not SMPS though