Test Equipment Anonymous (TEA) group therapy thread

[cyclin_al]

I bought some of those smd resistor/cap "kit" books...sadly it was summer & they used some cheapo extra gummy labels...By the time  it arrived the labels had all fallen off & were just a sticky clumpy mess in the package & the ink also "peeling off"....with all the approx 170 labels essentially missing, the book isn't so handy!

https://www.amazon.ca/gp/product/B08P8TYQ2S/

Well, that was kind of the reason I posted. When you look at these listings, they're an unknown; I figured a post showing what actually arrived for my 9 bux and change would help my fellow Canucks make a informed decision. Also, being 0805, they're quite a nice size for hand work. Hard not to get your money's worth here.  +--

EDIT: Woops! Looks like somebody took advantage... they've sold out the ones that were In Stock/Fulfilled by Amazon, and are now only shipping from China at CAD$45/copy. Hope those who partook enjoy their purchase!

mnem
*tsssssst...*

It may be possible that the last one in stock is now on its way to me; would need to confirm as SWMBO is the household fan of amazon ...

[mnementh]

Put it in perspective this way:

What do you imagine the guys who designed your hollow-state wonders would have thought of that meter...?

Especially if you told them it could be bought new for the price of steak dinner at a roadhouse...?

mnem
 

Well if it's not linear as Med suggests and the specification is only 3 times better than an AVO 8, probably not too much.

I knew I took too long to add that little bit... it really is very relevant here. Yeah, it's no FLUKE. But it doesn't have that price, or an AVO 8 price either.

As for the nonlinearity... agreed, that is a concern. But we don't know if that is a result of old age, abuse at the hands of a previous owner, or current owner user error.


The difference here is med has better, so he doesn't need to settle. But it is still more than good enough for most of the work most meters are used for.


mnem
 

[med6753]

Yep, but in reality, for what med does, it is perfectly usable, who has ever heard of Tektronix scope failing to function because the supply lines were 1mV adrift of their true voltage? Answer, nobody and nobody ever will.

Wanna bet? The Type 1A1 plug-in in the Type 547 is 1mV off vertically on both channels which is totally unacceptable and will be investigated and fixed.   

[Cerebus]

Considering that's the Trabant of DMMs, that's still fairly good 

I wonder if the boards are made out of recycled Duroplast from old Trabant panels? 

[Andrew_Debbie]

Someone wanted a 3478? This one has no bids yet, and who knows how the seller might react to a careful offer?

What is a reasonable price these days?    I had a new one on my desk at work many many years ago.  Fab meter when new.

 I'm hesistant to buy a used one because of the internal battery  and potential loss of calibration.   If it still has calibration data,   I can change the battery.   Recovering from lost cal data is not something I can do at home.   I've only got a 2.5000V reference and a 9999 count DMM.

The one I scored has date codes suggesting a build date around late 1986 - early 1987, and the battery measures 3.03v. Perhaps of more immediate concern are the 4 RIFAs...

If you really want one, getting a clean one for £100 seems reasonable, but I wouldn't be in any hurry to bid higher as there doesn't seem to be a shortage of these.

Thanks.       I'm watching the listing but I doubt I will bid.    I can wait for a better price.

[Cubdriver]

Reader's Digest version: High-voltage film capacitors are no longer available in "tall skinny can" format, only big rectangular boxes. How do I attach the replacement capacitor securely when the new cap's footprint is to large to fit on the PCB?

Unabridged version: I just bought an HP 6209B DC power supply (0-320V 0-100mA) off of the estuary of wickedness. It's a late model, and a real beauty - traces were taped by hand and gold-plated. I was looking forward to long, blissful nights with just me, a 12AX7, and a high voltage power supply. Well, it turns out that one of the caps is making noise, the kind that can be detected by ear. The seller was a good sport and, at my suggestion, gave me a partial refund. The user manual is available online ( http://www.nousnexus.com/Manuals/HP_6209B.pdf ) and helpfully contains a schematic and parts list. The cap that failed is C20 - "film 25μf 475Vdc". It's a radial and ~30mm in diameter (see image, attached). I hop on digi-key and all of the 25μF film caps rated for >= 475VDC are boxes with footprints like 42x35mm. How the heck do I fit a large footprint component on a PCB designed for a tall skinny cap can? I was thinking that maybe I could build a little stand out of unclad FR-4 and gobs of silicone, and then plop the cap on top so it's raised off of the PCB (see DaveCAD, attached). Of course I would make something more structurally sound than what I sketched, but you get the idea. Is there a better way to solve this problem? Glue the cap to the side of the enclosure and run wires? Use multiple smaller caps in parallel? Use an electrolytic cap? Build my own capacitor? Give up and take up golf instead?

I'm not sure the description in the manual is correct, it says film for C20 but when you search the military part sites for 0180-1848 it is then described as an aluminium electrolytic, see here; https://www.parttarget.com/5910-01-311-5312_5910013115312_PFP250YN2A1P2.html/-18280EA8-7F78-47CF-815F-0CF3C4CD45F8

When you look up the Sprague DFP (one of the alternate parts) it is also an aluminium electrolytic, I think the manual is incorrect, this wouldn't be the first error we have seen in old manuals.



David

Yeah, definitely an aluminum electrolytic - just looked at mine (happened to have the cover off as I was cleaning the range switch) and took a good look at it.  I’ve found Mouser to have a pretty good selection of radial electrolytics with higher working voltages.

-Pat

[Specmaster]

When is a Fluke not a Fluke? When it's a Mastech. Trying to calibrate the DCV is an exercise in futility. It is very non-linear across the DC range and getting an acceptable compromise is near impossible. Now it COULD be because I'm doing it wrong due to lack of service info. So I have to consider that. But for critical measurements this guy falls flat. So it will just be used to quickly check new stock capacitors.



That concludes all my calibrations for now except the Tek 2465 DMS with the DMM option. But that would involve lifting and that's a no-no right now. And the Type 547 sits there mocking me again. Now that it has good power I want to dive in and fix the remaining issues. 

Stop being so negative towards this meter, think you find it is actually pretty good, it is only 1mV out, and I think it will be well within its design specification, take at the data sheets below. I think I told mnem a lie, I think that the maker of these meters might well be Appa, so email them with a photo of yours and enquire there, you may just get lucky. Keep up the good behaviour and keep well clear of these heavy beasts that are mocking, they will still sitting there once you're mended, they aint going anywhere.   

Well it's easy to meet the spec, DC volts 0.3% +_5 counts 
This is supposed to be a 4.5 digit meter a mid range fluke 77 3.5 digit handheld does better, a 8010A is 0.1% _+1 count and a 8050A 4.5 digit does 0.03%

Edit, sorry it's 4000 count not 4.5 digit but still 0.3% +5 counts that is 

Yep, but in reality, for what med does, it is perfectly usable, who has ever heard of Tektronix scope failing to function because the supply lines were 1mV adrift of their true voltage? Answer, nobody and nobody ever will.

Fact is if this was the only meter he or anybody else ever had, it would be best meter for the job in their tool kits and it would not affect their ability to repair things in any way whatsoever. It is a massive jump up from an analogue meter and we all managed perfectly with those before digital became a reality. 

Well, regardless. I have 40 year old Flukes that run circles around it so in reality let's call it what it is. A POS. 

Geez, you're a real hard man to please, thats for sure. I sure would accept that meter just as it is, and I know that at the end of the day it's not going to make a single jot difference to any of the scopes you work on, especially seeing as a scope of your vintages at best has an accuracy of +/- 3% on readings, much the same as an analogue multimeter. In fact, I reckon as many of them, especially the hollow state ones were made, tested and even had their workshop manuals written up with analogue meter readings (20,000OPV) in mind. There will be many occasions where using any digital meter will not provide you with the correct readings as noted on the schematics purely because of the loading effect of an analogue meter of 20,000 OPV as opposed to a digital with 10,000,000 OPV input impedance. 

Put it into perspective, your Honda Civic in comparison to say a Cadillac Seville is a POS I dare say, but they both will get from point A to point B in reasonable comfort. Granted the Cadillac is several grades higher up the quality stakes and price stakes, just as the 40 years old Fluke is to the Mastech, they are not directly comparable. 

[duckduck]

Reader's Digest version: High-voltage film capacitors are no longer available in "tall skinny can" format, only big rectangular boxes. How do I attach the replacement capacitor securely when the new cap's footprint is to large to fit on the PCB?
<SNIP>

I'm not sure the description in the manual is correct, it says film for C20 but when you search the military part sites for 0180-1848 it is then described as an aluminium electrolytic, see here; https://www.parttarget.com/5910-01-311-5312_5910013115312_PFP250YN2A1P2.html/-18280EA8-7F78-47CF-815F-0CF3C4CD45F8

When you look up the Sprague DFP (one of the alternate parts) it is also an aluminium electrolytic, I think the manual is incorrect, this wouldn't be the first error we have seen in old manuals.

David

factory/David, you nailed it. I desoldered that cap and then chopped it open and, sure enough, it's got paper and thin aluminum sheet (the grey stuff) inside. It was dry as a bone in there.

EDIT: Also, cubdriver.

[Specmaster]

Yep, but in reality, for what med does, it is perfectly usable, who has ever heard of Tektronix scope failing to function because the supply lines were 1mV adrift of their true voltage? Answer, nobody and nobody ever will.

Wanna bet? The Type 1A1 plug-in in the Type 547 is 1mV off vertically on both channels which is totally unacceptable and will be investigated and fixed.   

Does it work? Yes it does, does the workshop manual state that the 1mV is critical? NO. You are making it critical because it does not sit comfortably with your OCD. I think that the figures quoted in the workshop manual have a fair degree of margins that as long as the voltages are within those parameters it is deemed to be fine. Even digital scopes unless paying thousands for them still have margins of around 3% for their digital readings on screen, which I wager are going to be far better than judging a reading against a graticule can ever be, even with perfect vision. 

[med6753]

Put it into perspective, your Honda Civic in comparison to say a Cadillac Seville is a POS I dare say, but they both will get from point A to point B in reasonable comfort. Granted the Cadillac is several grades higher up the quality stakes and price stakes, just as the 40 years old Fluke is to the Mastech, they are not directly comparable. 

No, the difference is that my Civic will get me to my destination. The Caddy, being a GM product, will blow up and leave me stranded. 

[med6753]

Yep, but in reality, for what med does, it is perfectly usable, who has ever heard of Tektronix scope failing to function because the supply lines were 1mV adrift of their true voltage? Answer, nobody and nobody ever will.

Wanna bet? The Type 1A1 plug-in in the Type 547 is 1mV off vertically on both channels which is totally unacceptable and will be investigated and fixed.   

Does it work? Yes it does, does the workshop manual state that the 1mV is critical? NO. You are making it critical because it does not sit comfortably with your OCD. I think that the figures quoted in the workshop manual have a fair degree of margins that as long as the voltages are within those parameters it is deemed to be fine. Even digital scopes unless paying thousands for them still have margins of around 3% for their digital readings on screen, which I wager are going to be far better than judging a reading against a graticule can ever be, even with perfect vision. 

Yes, the acceptable tolerance is 3%. But being 1mV low on a 5mV signal is certainly NOT 3%. 

[Specmaster]

Yep, but in reality, for what med does, it is perfectly usable, who has ever heard of Tektronix scope failing to function because the supply lines were 1mV adrift of their true voltage? Answer, nobody and nobody ever will.

Wanna bet? The Type 1A1 plug-in in the Type 547 is 1mV off vertically on both channels which is totally unacceptable and will be investigated and fixed.   

Does it work? Yes it does, does the workshop manual state that the 1mV is critical? NO. You are making it critical because it does not sit comfortably with your OCD. I think that the figures quoted in the workshop manual have a fair degree of margins that as long as the voltages are within those parameters it is deemed to be fine. Even digital scopes unless paying thousands for them still have margins of around 3% for their digital readings on screen, which I wager are going to be far better than judging a reading against a graticule can ever be, even with perfect vision. 

Yes, the acceptable tolerance is 3%. But being 1mV low on a 5mV signal is certainly NOT 3%. 

Granted, but do you have the means to be 100% sure that when that meter or the Fluke for that matter, when reading voltages as low as 5mV are going to be anywhere near correct unless we have a voltage reference that we can rely on at those levels. AD584-M will only provide us with 2.50V as the lowest. I have a Mastech calibrator that will provide output voltages from 1mV to 10V in 1mV steps, but I'm not at all sure how accurate it is I have no idea. At similar voltages to the AD584-M steps, they are very close, but which one is the best yard stick to use. Hence why I mentioned before about the differences of your Flukes on the AD584-M and your Siglent which does not agree with the Flukes, i.e., as I mentioned, they were both calibrated / setup against completely physically different references, so which one do you think was the most reliable. Given the age of the meters and the assumed huge différance in the cost of the references, my money would be the Siglent, but I can't be sure. It could even be down to the meter input resistance loading down the reference. By way of example, my Fluke 8840A and also 8842A and the TTI 1906 have input resistances on the mV ranges of over 50M, my other bench meters, Thurlby 1905A, Blackstar 4503 and HP2466A have resistances ranging from 10M to 12M, so measuring 5mV on those meters would provide differing results.

If I had to measure such low voltages, I think I would be looking for something specifically designed for that purpose, and also special leads, as most DMM's will pick up spurious signals in the air at such low values. 

[med6753]

I don't use the AD584-M's when checking scope inputs. I use a home built fixed square/triangle/sine calibrator plus a function generator. And the outputs are proven accurate and stable over dozens of years and dozens of scope calibrations. So I trust what it's telling me. And the Heath IG-4244 comes into play for compensation adjustments.

In the case of the Type 1A1 plug-in the inaccurate gain carries through in the upper ranges too. A 500mV p-p signal results in an 400mV p-p display. And the gain pot is topped out. The gain adjustment should allow me to go above 500mV p-p but it doesn't and that's clearly a defect.
And it's not a matter of incompatibility. The Type 1A1 can be used in an older Type 535A even though it has a 50Mhz B/W compared to the Type 535A 15Mhz B/W. 

And yes, I have OCD but if there's clearly a defect I'm going to do all I can to fix it. The above is a defect.     

[tggzzz]

A traditional technique for generating low voltages is to have a, say, 10V stable voltage source, and use a Kelvin Varley Divider to generate lower voltages. 7 decade KVDs are available, which gives microvolt resolution.

[BU508A]

I did a teardown of that Keithley 616 electrometer and mentioned, that I need a cable for it.
Which I've found in the bay, a Keithley 6011-10 (10ft long).

Well, the item has been delivered through USPS and now I understand completely med6753, why he is avoiding USPS.

Tracking history of said item (timeline is from bottom to top):

July 2, 2021, 9:46 am
Delivered, Front Desk/Reception/Mail Room
MINNEAPOLIS, MN 55428
Your item was delivered to the front desk, reception area, or mail room at 9:46 am on July 2, 2021 in MINNEAPOLIS, MN 55428.
July 2, 2021, 8:14 am
Out for Delivery
MINNEAPOLIS, MN 55428
July 2, 2021, 4:46 am
Arrived at Post Office
MINNEAPOLIS, MN 55429
July 2, 2021, 3:05 am
Arrived at USPS Facility
MINNEAPOLIS, MN 55429
July 2, 2021, 2:52 am
Departed USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 10:22 pm
Arrived at USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 10:07 pm
Arrived at USPS Facility
MINNEAPOLIS, MN 55413
July 1, 2021, 9:23 pm
Arrived at USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 7:57 pm
Departed USPS Facility
SAINT PAUL, MN 55131
July 1, 2021, 8:00 am
Arrived at USPS Facility
SAINT PAUL, MN 55131
July 1, 2021, 7:44 am
Arrived at USPS Regional Facility
SAINT PAUL MN NETWORK DISTRIBUTION CENTER
July 1, 2021, 7:04 am
Departed USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 2:54 am
Arrived at USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 2:11 am
Departed USPS Facility
SAINT PAUL, MN 55131
June 30, 2021, 10:59 pm
Arrived at USPS Facility
SAINT PAUL, MN 55131
June 30, 2021, 5:20 pm
Departed Post Office
OSSEO, MN 55369
June 30, 2021, 5:12 pm
Acceptance
OSSEO, MN 55369
June 30, 2021, 9:31 am
Shipping Label Created, USPS Awaiting Item
MINNEAPOLIS, MN 55428 

I've contacted the seller, and this is the outcome so far:
BU508A:
Hello,

I've checked today the tracking information and I've found, that the item has been delivered.
Only problem is: item has been delivered to some place in Minnesota but I am living in Germany.
I assume, that the tracking ID is wrong. Could you please provide me the correct tracking ID?
Wrong Tracking ID (USPS): LG00XXXXXXXUS
https://tools.usps.com/go/TrackConfirmAction_input?strOrigTrackNum=LG00XXXXXXXUS
Thank you for your help.
Best regards,
BU508A

plymouthsales4165:
For some reason the post office delivered it back to us so we dropped it back at the post office to go to you

BU508A:
Is there a new tracking ID which you can provide me?

plymouthsales4165:
It is the same tracking number. The post office said that since it was miss directed and sent back to us that it could be reshipped on the same label

BU508A:
Okay, I've checked the tracking ID but it shows still the same result as "delivered". I will check in one day again, if there is an update. Thanks for your answers.

plymouthsales4165:
Sounds good

I'm really curious how this story will end.

[bd139]

They must have shipped you a boomerang by accident?

[BU508A]

This crossed my mind, yes. 
But I hope, that this boomerang will hit finally my place AND turn out to be a Keithley cable with a TRIAX connector. 

[Cerebus]

...
July 2, 2021, 2:52 am
Departed USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 10:22 pm
Arrived at USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 10:07 pm
Arrived at USPS Facility
MINNEAPOLIS, MN 55413
July 1, 2021, 9:23 pm
Arrived at USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 7:57 pm
Departed USPS Facility
SAINT PAUL, MN 55131
July 1, 2021, 8:00 am
Arrived at USPS Facility
SAINT PAUL, MN 55131
July 1, 2021, 7:44 am
Arrived at USPS Regional Facility
SAINT PAUL MN NETWORK DISTRIBUTION CENTER
July 1, 2021, 7:04 am
Departed USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 2:54 am
Arrived at USPS Regional Facility
MINNEAPOLIS MN DISTRIBUTION CENTER
July 1, 2021, 2:11 am
Departed USPS Facility
SAINT PAUL, MN 55131
June 30, 2021, 10:59 pm
Arrived at USPS Facility
SAINT PAUL, MN 55131

So they drove it over the Mississippi and back again, and then back again?    (For those not in the know, Minneapolis-St Paul does the same trick as Budapest, it's one city that calls itself different things depending on which side of the river you're on.)

[med6753]

Well at least USPS didn't throw the package INTO the Mississippi River when going back and forth. 

USPS sucks donkey balls. 

[vk6zgo]

Put it in perspective this way:

What do you imagine the guys who designed your hollow-state wonders would have thought of that meter...?

mnem
 

Well if it's not linear as Med suggests and the specification is only 3 times better than an AVO 8, probably not too much.

I bet it doesn't have a 2.5kv (later increased to 3kv) range on both DC & ac like an AVO 8, though!

[vk6zgo]

Yep, but in reality, for what med does, it is perfectly usable, who has ever heard of Tektronix scope failing to function because the supply lines were 1mV adrift of their true voltage? Answer, nobody and nobody ever will.

Wanna bet? The Type 1A1 plug-in in the Type 547 is 1mV off vertically on both channels which is totally unacceptable and will be investigated and fixed.

That's what the Good Lord gave us variable controls for!

Back in the day, at my old work, the tame EEs produced an excellent TV transmitter testing manual.

Engineers are lovely folk, but they are wont to over complicate things, so that if the measurement was supposed to be in %, they read off, as well as they could from an analog graticule, the values of both the reference voltage & the measured voltage, then calculated the percentage value.

Techs tried this, recorded the resultant values, compared them to the logical way they had been doing it for years, which was:-

Using variable v/div, set the reference to 5 major divisions, then read the value being measured off directly in % .(for small values of the latter, the x2 & x5 step switch positions still work in this mode)
Not enough difference to matter a damn!

One guess which method was used from then on!

One weird thing, though, was that some "%" readings with analog video testing don't refer to actual percentages of the tested values, but to the percentage of people who could notice a particular picture error in a "double blind" test.

These were tested for with special graticules, again requiring the use of variable v/dv.

[Specmaster]

I don't use the AD584-M's when checking scope inputs. I use a home built fixed square/triangle/sine calibrator plus a function generator. And the outputs are proven accurate and stable over dozens of years and dozens of scope calibrations. So I trust what it's telling me. And the Heath IG-4244 comes into play for compensation adjustments.

In the case of the Type 1A1 plug-in the inaccurate gain carries through in the upper ranges too. A 500mV p-p signal results in an 400mV p-p display. And the gain pot is topped out. The gain adjustment should allow me to go above 500mV p-p but it doesn't and that's clearly a defect.
And it's not a matter of incompatibility. The Type 1A1 can be used in an older Type 535A even though it has a 50Mhz B/W compared to the Type 535A 15Mhz B/W. 

And yes, I have OCD but if there's clearly a defect I'm going to do all I can to fix it. The above is a defect.     

Yes, I agree its a defect, but most scopes that I have seen have extremely noisy floors at the low mV levels anyway, and voltages that low really need meters designed for that level of input, not general purpose bench meters.

I was referring to any of the power supply rails, be they 3.3V, 5V, 7.5V, 15V, 30V or 100V, or even transistor biasing voltages then any of your current meters would suffice to verify those. So if the 5V power rail was 4.99V or 5.01V, then that would not cause the issue that you have.

[tautech]

Always getting  on Discord for not having a working Mic so prior to delivering a new scope to a large audio crowd some horse trading was arranged and access to their E-Waste room was approved. 

Some fixing first might be required however but still quite pleased with the score.
Surely I can get one of the Mic I/O's working. 
Now to look for manuals etc 
Found in a jiffy:
https://www.soundcraft.com/en/product_documents/notepadusermanual_v-1-0-pdf

The continuing saga of the E-waste Soundcraft Notepad 12FX.....
As reported the 18V supply into the rear barrel jack had a short and while some parts of 18V supply to the PCB could be disconnected with 0R 0805 jumpers the numerous switching supplies for 48, +15, 5 and 3.3V can't as as it's all SMD and not happy to dig deeper until a service manual or schematic was found. 

A wee pic of what I'm up against minus a few bits desoldered for better access:


So a weekend email to the chap that kindly donated this stuff to tautech was returned before midday Monday with a full set of schematics. 
Who's ever heard of that sort of support for something from an E-waste bin ? 

Just what I need other than more spare time to progress further:

[mnementh]

A Fabricobbledy Monday Morning Project:



The resistors were waiting in my mailbox this morning after brekkerz; I decided to step back from that Pioneer to get a start on something useful, so pulled the heatsink from that black behemoth monoblock amp to make this. 20 m3 holes drilled and tapped with the same bit/tap, all by hand with cordless drill. Even had to tap the holes with the drill too... Okay, I didn't have to, but I preferred to as I didn't have a proper ratcheting tap handle to use.

Still to come: Terminals on the way, and designing/printing bits to hold them and a 120mm fan...   

And here we are, Monday morning again, just finishing this up. Started out yesterday morning designing and printing the 4 angle plates.



While they printed, I made and soldered up the plumbing. Yes, those are recycled bits of ATX power supply wires. 



When it came time to dry-fit, I realized I wanted the body of the spring terminals to be under the plate; a couple minutes in Frustion360 yielded these plates which printed up in aboot 20 minutes. They're designed to self-thread for m2.5, as I have a MFT of them I bought by accident and never use. Only aboot 190 more to go... 



The endplate is actually a remix of the MSI dragon rendered as a fan grille; I stretched it to 6mm thick in Cura so I could have some decent rigidity. The bracket with feet was a bit of a PITA to design, but I did finally beat Frustion360 into submission. It holds the fan ~30mm above the surface the stack rests on, so the whole thing is pretty much a fan-assisted chimney.

So what is it...? I'll let y'all guess. Hints: Input is on the right, and I'm intending to push each set of resistors at 30-40W total.

Oh, and it's something I've been meaning to make literally for decades but kept putting off until now, when it became important because of WAF.

Now I know y'all will have misgivings aboot something like this actually dissipating 150-200W and being made almost entirely of low-temp thermoplastic. I do as well, to some extent. However, this whole stack sits quite solidly on top of that fan without even having a frame around it; I'm reasonably certain that it's going to be fine for its intended purpose, as it will never be used for anything but low-voltage loads.

mnem
*knocks self unconscious with a decibel mallet*

[ch_scr]

Now I know y'all will have misgivings aboot something like this actually dissipating 150-200W and being made almost entirely of low-temp thermoplastic. I do as well, to some extent. However, this whole stack sits quite solidly on top of that fan without even having a frame around it; I'm reasonably certain that it's going to be fine for its intended purpose, as it will never be used for anything but low-voltage loads.

mnem
*knocks self unconscious with a decibel mallet*

At least you'll have the files ready to print it from some material better suited