Test Equipment Anonymous (TEA) : Discussion and Group Therapy Thread

[mnementh]

It's a Dell fleet box... first thing I'd check is the PSU

dere u go (do you hear my Italian accent?)

2022-11-30 00_05_56-Window.jpg. 2022-11-30 00_08_54-Window.jpg. 2022-11-30 00_09_36-Window.jpg.

looks ok. Crappy fan.... 2022-11-30 00_10_05-Window.jpg

let's see if it boot first when I get the RAM.

Looks pretty clean and probably low hours for a fleet box. No visibly leaking or bloaty caps. But it really is the load test that's important. These commodity-grade PSUs are notorious for getting noisier and noisier as load increases - even brand new outta da box.

I'd also try booting with only a single DIMM until you have a chance to run it a while; that way you're not risking a whole $150ish kit of RAM on a unknown PSU and MB/VRMs.

mnem
err on the side of caution.

[mnementh]

Audio stuff is like eating chocolate, even if it is not necessary is such a joy...

Yeah... it's especially fun to find hidden/unknown (well, I didn't know what a treasure it is, until I had one) treasures like this one. It is so convenient to be able to just clip it to the collar of my t-shirt, and I still get 4-5 hours on a charge (design life is 20ish), and I'm really loving the fact that I can turn it up loud enough that I can't hear the kids, and especially that I can listen to my music without the the fucking phone interrupting it for some stoopit eMail or SMS notification.

mnem
Sometimes a "single purpose device" is better.

[Zucca]

I'd also try booting with only a single DIMM until you have a chance to run it a while; that way you're not risking a whole $150ish kit of RAM on a unknown PSU and MB/VRMs.

$180 64GB RAM already gone, well parked... Amazon can refund me in 30days window... will keep you posted.... only 460W PSU.... should be enough... if not I could buy something with modular cables, that one has still all the spaghetti coming out the PSU and vomited in the mini tower...

[Cubdriver]

The THD+N at 1000 hz at 1 volt (top) and 5 volts (bottom) isn't catastrophically bad but not as good as it could be, I think, so I'm going to let it run overnight and see if or how it changes by lunchtime tomorrow per the suggestions in one of Simon Spier's videos while I work on the other stuff I have going.

My HP 8903B readings after not quite 12 hours on have changed slightly. No filter readings are the same, or slightly worse, and filtered readings have improved a bit.

Initial Current Delta Configuration
0.0068 0.0071 +0.0003 1 kHz @ 1 V, no filter
0.0028 0.0024 -0.0004 1 kHz @ 1 V, 80 kHz low pass
0.0038 0.0038 0 1 kHz @ 5 V, no filter
0.0023 0.0019 -0.0004 1 kHz @ 5 V, 80 kHz low pass

-Pat

Eidt - in my draft, the above was spaced out in a nice table. Apparently the formatting doesn't carry over to the actual post.

[25 CPS]

The THD+N at 1000 hz at 1 volt (top) and 5 volts (bottom) isn't catastrophically bad but not as good as it could be, I think, so I'm going to let it run overnight and see if or how it changes by lunchtime tomorrow per the suggestions in one of Simon Spier's videos while I work on the other stuff I have going.

My HP 8903B readings after not quite 12 hours on have changed slightly. No filter readings are the same, or slightly worse, and filtered readings have improved a bit.

Initial Current Delta Configuration
0.0068 0.0071 +0.0003 1 kHz @ 1 V, no filter
0.0028 0.0024 -0.0004 1 kHz @ 1 V, 80 kHz low pass
0.0038 0.0038 0 1 kHz @ 5 V, no filter
0.0023 0.0019 -0.0004 1 kHz @ 5 V, 80 kHz low pass

-Pat

Eidt - in my draft, the above was spaced out in a nice table. Apparently the formatting doesn't carry over to the actual post.

Don't worry, I was still able to understand it. The figures on mine crept up slightly too as of last check but part of it might be the GPS disciplined oscillator testing I'm doing right in front of it adding electrical noise. Not from the 10 MHz stuff in itself but there are three Chinese switch mode adapters powering the GPSDO and the two distribution amplifiers within arms's reach of the 8903B.

[mansaxel]

Wifey & kids keep snurching my medium & small spring clamps to put on cereal & chip bags... and then they vanish. I literally go to Horror Fraught every few months to buy more.

mnem
clamps.jpeg

Here, we've got special Devices for this:



This is IKEA BEVARA ("keep") bag clamps. Get a couple bags of these. They can clamp a plastic bag air tight without breaking a sweat.

[mansaxel]

Thanks for posting that - it's reassuring to see some consistency between the performance of the two HP 8903Bs. I was playing around with the 80 kHz filter on mine while I was watching the GPS disciplined oscillator results and I saw a similar reduction in THD+N figures when it's in circuit.

I just finished posting the GPSDO topic and it's got lots of pictures including one showing both burn in tests set up to go overnight. If all goes well, it looks like I'll be stopping in at the local electronics brick and mortar tomorrow to pick up some cables so I can start wiring in 10 MHz reference permanently.

I just saw your GPSDO post. I *still* haven't set up the one I bought years ago - still need to rig an antenna. Probably not gonna be doing it now with the cold weather coming. Those DAs look interesting.

I'll leave my 8903B on overnight to see if anything changes, too, so there will be more data points.

-Pat

Meinberg thinks my GPSDO needs a new OXCO. Thinking about bringing the Anritsu MT1000A home for a frequency check. And then I hope I can snatch and repurpose some of those CTC "sawn out of base stations" oscillators for that.

[25 CPS]

I heard the mail box flap followed by the screen door a little bit ago indicating Canada Post came again and dropped off another box.



This is another HP 35 that I ordered as a parts unit in case I needed one to get the first one working. This 35 works but it looks like the power adapter might be bad. I'm going to have to get out some fine probes to check the output of the two suspect adapters. Right now, I can power up either 35 or the 45 using the adapter that came with the 45. I tried this 35's adapter on it, the other 35 and the 45 without any success.

I was thinking the only thing I didn't have with that first HP 35 was the original bill of sale but when I was working through the manual that came with the first HP 35 the other night, the receipt fell out of the folder pocket inside the back cover of the manual along with two unused "Property of" name tag stickers. I'll update the HP 35 thread with more pictures and results from checking over adapters.

[mansaxel]

Won the Cropico milliohm meter I talked about earlier. Classic snipe operation. 0 bids up to 3 minutes left, then someone left a starting bid; at -30 seconds a second bid came through, and I shot the competition to pieces at -3 seconds.

(pic from ad)



I'll post some more pics of it as I have it.

Why did I want it? Well, first and foremost, it is TE. That in and of itself, and so on.

I also wanted to have a way to measure very low resistances. Like across a 221 quick connect, and comparing it to the higher? resistance of a wire nut.

[Zucca]

Remember a milliohm meter is like a microscope, you will see stuff you won't believe is there.
Nice deal.... one day I will have one too....

[mnementh]

Wifey & kids keep snurching my medium & small spring clamps to put on cereal & chip bags... and then they vanish. I literally go to Horror Fraught every few months to buy more.

mnem
clamps.jpeg

Here, we've got special Devices for this:

This is IKEA BEVARA ("keep") bag clamps. Get a couple bags of these. They can clamp a plastic bag air tight without breaking a sweat.

But then I'd be going to IKEA every few months to replace the ones that vanish...

mnem
Maybe I could just leave the wife & kids at IKEA...

[mansaxel]

This is IKEA BEVARA ("keep") bag clamps. Get a couple bags of these. They can clamp a plastic bag air tight without breaking a sweat.

But then I'd be going to IKEA every few months to replace the ones that vanish...

If you buy enough, and mandate their return to the kitchen after use, you'll be able to maintain a collection. They're sturdy and hold up to use well. I think we have about 100 or so, and there usually are a couple ready to use in the kitchen drawer.

[Robert]

Won the Cropico milliohm meter I talked about earlier. Classic snipe operation. 0 bids up to 3 minutes left, then someone left a starting bid; at -30 seconds a second bid came through, and I shot the competition to pieces at -3 seconds.

(pic from ad)



I'll post some more pics of it as I have it.

Why did I want it? Well, first and foremost, it is TE. That in and of itself, and so on.

I also wanted to have a way to measure very low resistances. Like across a 221 quick connect, and comparing it to the higher? resistance of a wire nut.

Nice meter. It's a DC measurement one IIRC. I have a few low resistance meters. In aviation they are often called bonding meters as they are used to check the resistance of electrical "bonding" used for lightning protestion. Some use AC as this revmoves errors caused by thermal or electrochemical EMFs between dis-similar metals.

[25 CPS]

The THD+N at 1000 hz at 1 volt (top) and 5 volts (bottom) isn't catastrophically bad but not as good as it could be, I think, so I'm going to let it run overnight and see if or how it changes by lunchtime tomorrow per the suggestions in one of Simon Spier's videos while I work on the other stuff I have going.

My HP 8903B readings after not quite 12 hours on have changed slightly. No filter readings are the same, or slightly worse, and filtered readings have improved a bit.

Initial Current Delta Configuration
0.0068 0.0071 +0.0003 1 kHz @ 1 V, no filter
0.0028 0.0024 -0.0004 1 kHz @ 1 V, 80 kHz low pass
0.0038 0.0038 0 1 kHz @ 5 V, no filter
0.0023 0.0019 -0.0004 1 kHz @ 5 V, 80 kHz low pass

-Pat

Eidt - in my draft, the above was spaced out in a nice table. Apparently the formatting doesn't carry over to the actual post.

I typed out a whole long response on my smartphone and all it submitted was the quote. Pardon me while I start all over again.

[25 CPS]

The THD+N at 1000 hz at 1 volt (top) and 5 volts (bottom) isn't catastrophically bad but not as good as it could be, I think, so I'm going to let it run overnight and see if or how it changes by lunchtime tomorrow per the suggestions in one of Simon Spier's videos while I work on the other stuff I have going.

My HP 8903B readings after not quite 12 hours on have changed slightly. No filter readings are the same, or slightly worse, and filtered readings have improved a bit.

Initial Current Delta Configuration
0.0068 0.0071 +0.0003 1 kHz @ 1 V, no filter
0.0028 0.0024 -0.0004 1 kHz @ 1 V, 80 kHz low pass
0.0038 0.0038 0 1 kHz @ 5 V, no filter
0.0023 0.0019 -0.0004 1 kHz @ 5 V, 80 kHz low pass

-Pat

Eidt - in my draft, the above was spaced out in a nice table. Apparently the formatting doesn't carry over to the actual post.

Don't worry, I was still able to understand it. The figures on mine crept up slightly too as of last check but part of it might be the GPS disciplined oscillator testing I'm doing right in front of it adding electrical noise. Not from the 10 MHz stuff in itself but there are three Chinese switch mode adapters powering the GPSDO and the two distribution amplifiers within arms's reach of the 8903B.

Let’s try this again…

By the time I got downstairs again and checked everything, it had been close to about 20 hours of running time. Neither the 8903B or the GPSDO had moved from where they were when I last checked on them last night and took the pictures I posted late night/early morning. The good news is they have been stable.

I don’t know what effect the three Chinese switch mode power supplies right in front of the 8903B had on the THD+N it was registering once I powered up the GPSDO stuff since it’s possible they could be adding to the +N being measured.

That has given me some more thoughts on installation. I’m going to keep the three adapters on their own power bar, probably the one they’re on now not only because of the physical clearance issue with the other outlets arising from how the plugs and cases were laid out. But What I’m thinking of doing is mounting the HSR to the shelves on the back wall on the other side of the walkway area behind the bench shelves and running the length of the DC cables to the GPSDO and distribution amplifiers instead of bringing the AC power (and consequently the adapters) to them, to get a bit of physical separation between them and the test equipment.

I just cleared out the local Sayal Electronics store of most of their RG-58U cables. This wasn’t too hard with how low their stock was. That’ll let me start wiring everything up and installing this properly once I get home from doing a few other errands after I finish my coffee. Unfortunately, I’ve got a shift rotation at work starting up so after this evening, free time is going to be difficult to come by for the next few weeks. I was really enjoying spending some quality time down in the workshop since I had the time off and there were no big house related projects that needed to be done. And just like that, it’s going to be over for the most part for several weeks.

[mnementh]

I'd also try booting with only a single DIMM until you have a chance to run it a while; that way you're not risking a whole $150ish kit of RAM on a unknown PSU and MB/VRMs.

$180 64GB RAM already gone, well parked... Amazon can refund me in 30days window... will keep you posted.... only 460W PSU.... should be enough... if not I could buy something with modular cables, that one has still all the spaghetti coming out the PSU and vomited in the mini tower...

24V Load Test.jpg

I was thinking an actual test OOC, looking with a scope to the noise level of the PSU with at least a moderate (20-50% of max spec) load on it, for the 12V and 5V rails. Even if you have to resort to something a little crude to get yer load on...

TL/DR version is I don't trust those little boxes of fire as far as I can fling 'em.

mnem
"Here you go... catch!!!"

[Zenith]

TL/DR version is I don't trust those little boxes of fire as far as I can fling 'em.

At one time ATX PSUs could be bought really cheaply, around £8 for a 500W PSU. I found they lasted about a year. No flames and smoke, but they either just stopped working or played up in strange ways, such as the PC crashing when the CD tray was opened. On opening them up them they had bulging caps with never heard of brands.

Later I started spending sensible money (£50-60) and bought Corsair. I've never had one fail and they clock up a lot of hours. I've had a fan become noisy but it was OK after it had been opened up and re-oiled. It pays top open them up every not and then and blow the dust out.

For makers of pre-built PCs, PSUs must be a tempting target for corner cutting. Dell always had a reputation for cutting the spec on their power supplies right to the limits.

[mansaxel]

Nice meter. It's a DC measurement one IIRC. I have a few low resistance meters. In aviation they are often called bonding meters as they are used to check the resistance of electrical "bonding" used for lightning protestion. Some use AC as this revmoves errors caused by thermal or electrochemical EMFs between dis-similar metals.

I have a Gossen one too, mainly for earth resistance measurements using spikes. Lots of dissimilar in that application. They opted for AC derived from a Zerhacker (Vibrator) in that one. Believe this one is a bit more sensitive and easy to use.

[Cubdriver]

My HP 8903B readings after not quite 12 hours on have changed slightly. No filter readings are the same, or slightly worse, and filtered readings have improved a bit.

Initial Current Delta Configuration
0.0068 0.0071 +0.0003 1 kHz @ 1 V, no filter
0.0028 0.0024 -0.0004 1 kHz @ 1 V, 80 kHz low pass
0.0038 0.0038 0 1 kHz @ 5 V, no filter
0.0023 0.0019 -0.0004 1 kHz @ 5 V, 80 kHz low pass

-Pat

As another data point, 24 hours later the readings remain basically unchanged from this time yesterday.

IIRC there was a big thread years back on the DIY Audio forum about improving the performance of an HP audio analyzer, but I don't remember if it was the 8903 or the 339A. Regardless, I think 0.00X figures are good enough for my purposes. Nice to see it's stable over a relatively long period of operation.

-Pat

[Zucca]

TL/DR version is I don't trust those little boxes of fire as far as I can fling 'em.

For makers of pre-built PCs, PSUs must be a tempting target for corner cutting. Dell always had a reputation for cutting the spec on their power supplies right to the limits.

Got it, but I will not exchange or test the PSU at this point. First the boot test, then see if it runs W10 stable (upgrade BIOS and run stress tests), then we will see if I have any bad PSU symptoms.
There will be no power hungry devices connected anyway, i.e. no big video cards I will use the integrated graphic chips in the MB.

So I am confident I will not push that unknown PSU anywhere close to its limits.

[Specmaster]

Rant mode on.
As many will already know, I'm busy decorating the hall, stairs and landing and trying to get all done before Christmas. That is not now going to happen, because of the sheer size of the area and the amount of white gloss paint already on the doors and stairs etc, it is taking an absolute age to remove the white paint from the stairs alone, so we can return the handrail to the varnished hardwood it was when we moved in 32 years ago. This means that we have to use a mulitool with a triangle sanding head to get in-between the spindles to sand back the spindle to prepare a smooth surface for painting. We have already electric heat gunned the old paint to scrap it off, which also was a real chore

The sanding heads and pads are of the hook and eye type, and they are so infuriating as the pads slip off the heads after a few minutes as the hooks on the heads give up and lose their grip. Now I know that many people simply love to hate Lidl tools, but my multitool is one of theirs, it's a Parkside and I've had this tool for years and years and the sanding head has finally given up the ghost and I have since purchased replacements from Bosch and others from Amazon, the Bosch came from B&Q and the sanding pads simply refuse to attach to it FFS, so that went for a refund and others ordered from Amazon, ordered 2, both failed after 30 to 40 minutes in total usage, complained to the Amazon seller who is sending me replacements. I also ordered other makes of heads, all are useless, so the original Parkside head was far superior, but replacements from Parkside are not obtainable

Searched around on Google, and it seems this is a common mode failure and others claim to have found a solution, put the head on the tool and switch it on and then apply it to a sheet of sandpaper and sand off the hooks until you get a smooth surface and apply some genuine self-adhesive Velcro hooks in 100 mm wide tape to the smooth head. This they claim works a treat, and it is then possible to use all the abrasive pad before having to replace it because of the lack of grip.

Going to order some of this tape from Amazon and see if its true, just want to get this decorating over and done with, Geez.

Has anyone else had similar problems and overcome it?

[Robert]

No problem with my DeWalt oscillating multi-tool and velcro sandpaper.

[tautech]

Rant mode on.
As many will already know, I'm busy decorating the hall, stairs and landing and trying to get all done before Christmas. That is not now going to happen, because of the sheer size of the area and the amount of white gloss paint already on the doors and stairs etc, it is taking an absolute age to remove the white paint from the stairs alone, so we can return the handrail to the varnished hardwood it was when we moved in 32 years ago. This means that we have to use a mulitool with a triangle sanding head to get in-between the spindles to sand back the spindle to prepare a smooth surface for painting. We have already electric heat gunned the old paint to scrap it off, which also was a real chore

The sanding heads and pads are of the hook and eye type, and they are so infuriating as the pads slip off the heads after a few minutes as the hooks on the heads give up and lose their grip. Now I know that many people simply love to hate Lidl tools, but my multitool is one of theirs, it's a Parkside and I've had this tool for years and years and the sanding head has finally given up the ghost and I have since purchased replacements from Bosch and others from Amazon, the Bosch came from B&Q and the sanding pads simply refuse to attach to it FFS, so that went for a refund and others ordered from Amazon, ordered 2, both failed after 30 to 40 minutes in total usage, complained to the Amazon seller who is sending me replacements. I also ordered other makes of heads, all are useless, so the original Parkside head was far superior, but replacements from Parkside are not obtainable

Searched around on Google, and it seems this is a common mode failure and others claim to have found a solution, put the head on the tool and switch it on and then apply it to a sheet of sandpaper and sand off the hooks until you get a smooth surface and apply some genuine self-adhesive Velcro hooks in 100 mm wide tape to the smooth head. This they claim works a treat, and it is then possible to use all the abrasive pad before having to replace it because of the lack of grip.

Going to order some of this tape from Amazon and see if its true, just want to get this decorating over and done with, Geez.

Has anyone else had similar problems and overcome it?

Yep, done lots and lots of this.
These hook/loop Velco sanding pads don't like any significant pressure/force on them of they will spit the paper off in short order.
Once the pads get tired stripping the velco from them and adding new stuff is only a stop gap measure and WILL let you down.
Instead find new pads in the style and thread type you need on Ali or ePay and get a pair or more. I've taken to getting sanding pads from them too. Stay with coarse papers initially, I only use 40 grit for outside work so to give primer the best possible mechanical key and prefer to add additional coats with sanding between to get a really smooth finish rather than sand fine in the first instance.

Heat gun stripping with a 2" broad knife done miles of too, 70 years of paint on a 200m2 weatherboard house all down to bare wood and what didn't come off did with a 7" pad with 24 grit on a 9" grinder ! Then smooth up with rotary orbital using 40 grit and start applying paint, 4 layers seals things up pretty good.

[25 CPS]

My HP 8903B readings after not quite 12 hours on have changed slightly. No filter readings are the same, or slightly worse, and filtered readings have improved a bit.

Initial Current Delta Configuration
0.0068 0.0071 +0.0003 1 kHz @ 1 V, no filter
0.0028 0.0024 -0.0004 1 kHz @ 1 V, 80 kHz low pass
0.0038 0.0038 0 1 kHz @ 5 V, no filter
0.0023 0.0019 -0.0004 1 kHz @ 5 V, 80 kHz low pass

-Pat

As another data point, 24 hours later the readings remain basically unchanged from this time yesterday.

IIRC there was a big thread years back on the DIY Audio forum about improving the performance of an HP audio analyzer, but I don't remember if it was the 8903 or the 339A. Regardless, I think 0.00X figures are good enough for my purposes. Nice to see it's stable over a relatively long period of operation.

-Pat

That's pretty much what I experienced with mine - after a lengthy warmup, it didn't move around anymore. It's reassuring to see this behaviour is consistent across multiple machines too.

I'm going to have to go looking over at the DIY Audio forum. I've never been on it so I'm curious to see what's been written about the 8903 and other distortion analyzers in general over there, and see if any of the people from one of the audio forums that I used to be active on have moved over there.

[Zucca]

This 4 wire meas explanation

https://learn.adafruit.com/adafruit-max ... ds-2928771

is not what I was expecting (I mean)
Datasheet MAX31865

https://cdn-learn.adafruit.com/assets/a ... X31865.pdf

bah....

Anyone with experience with PT100 and PT1000 in 180-450C° range? What is best PT100 or PT1000 if both in 4 wire setup?
Learning mode on...