EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Martin72 on July 15, 2023, 09:52:26 pm
-
Hi,
Start the thread anyway.
Here we go...
To keep the other thread (https://www.eevblog.com/forum/testgear/dmmcheck-plus-multimeter-calibration-reference-experiences/msg4963324/#msg4963324) cleaner.
Stumbled over this a few days ago:
https://www.elektormagazine.de/labs/voltage-current-calibrator-0-to-10v-and-0-to-40ma-0001 (https://www.elektormagazine.de/labs/voltage-current-calibrator-0-to-10v-and-0-to-40ma-0001)
https://www.youtube.com/watch?v=i8ZxOXWNTnU (https://www.youtube.com/watch?v=i8ZxOXWNTnU)
Looks really neat..
Already ordered the PCB, then look into the BOMs....Boy...not a cheapo. ;)
JLCPCB also offer 3d printing even in metal (interesting for the frontpanel and the heater housing)and so damn cheap, that the one, two expensive components (one chip, then the display) won´t hurt so much.
10ppm accuracy sounds not too bad, for the calibration of the calibrator a voltmeter with at least 6.5 digits is recommended.
When I decide to build this I would let it calibrate, not by myself.
-
Here a (temporary, will delete it in 24h) link to the docs:
-
Ooh, I'm happy to see the KiCad files. That means I can run DRC and mess around a little before ordering.
-
Nice...
Maybe I change to KiCad too, my premium license for Eagle is void.
-
I love KiCad. It works great, and it's free. 😉
I've also used FreeCAD to make 3D objects for my PCB projects. Like the largest inductor on my PCB in the other thread.
-
I'm cleaning up the silkscreens. C12 and C14 are labeled backwards. This is somewhat obvious if you pay attention to the fact that C14 is polarized, and the silkscreen next to C12 is polarized. They're parallel to each other, so it doesn't really matter, but it's pretty sloppy on their part.
-
Much better. Now it's accurate and easier to read. 👍
-
For whatever reason JLPCB has problems with the correct display of the placement imprint anyway.
So the designations like to "disappear" into the component print, when I think of it, I always position the designations with some distance to it.
This time I have not thought about it, because not my board.. ;)
-
I've been using JLC for years, and I haven't really had any trouble with that. This design had the silkscreens literally on top of pads and other stuff that would make it useless anyway. The heater board was worse, but I fixed that too.
Right now, I'm cleaning of the tracks. It's a real mess. It had pretty close to 100 errors in DRC. Nothing terrible yet, mostly just really sloppy traces going nowhere for no reason in random spots.
Edit: What thickness did you order for these PCBs?
-
Always the standard, 1.6mm .
It´s more than enough as you don´t have any heavy weight parts on it.
For a e.g. transformer board we use 2.4mm, anything less than this let it all hang down. ;)
-
Interesting, the 3d printing parts were rejected from JLCPCB, reason:
Too thin, they recommend at least 1.5mm thickness.
No problem, I must calculate the whole thing, because:
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-2-mini-detail (https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-2-mini-detail)
Con: No current source, Pros: More precise, (inhouse-) calibrated on a 8.5 digit meter...
And it´s complete assembled, tested, for 325 pounds.
-
Con: No current source, Pros: More precise, (inhouse-) calibrated on a 8.5 digit meter...
And it´s complete assembled, tested, for 325 pounds.
Knowing you, it's going out for calibration either way. 😉
The DIY version looks like a fun project. If I make it, I'll use my laser to cut a clear panel for it. I can also mark the outside of the enclosure for the cuts. I eventually want a fiber laser, so if I get that first, that can make the enclosure cuts too.
Edit: It's a touch screen. I'll probably just put a screen protector on it.
I wouldn't be surprised if you went for both projects anyway. Look at it this way, you can get the PDVS2mini now, and then wait until next year when the parts are cheaper to build the project reference. That practically pays for the PDVS2.
...and then buy another reference you find somewhere else. 🤣
-
At work, I am the testfield manager.
At home, I cultivate my hobby, which also happens to have something to do with electronics.
Then there's the fact that I'm curious and open to everything that goes in that direction.
And I like to pay apprenticeship money, because as ALF once said:
There is nothing good, unless you do it. ;)
-
ALF also said: “Where do you keep your casserole dishes? The cat wont fit in the toaster.”
-
I made a project on Mouser with all the main board parts. It's $366 USD including the $85 version of that out of stock chip.
I attached a Mouser BOM, but the parts need to be verified. The original supplied BOM has accuracy issues with the reference numbers.
-
Here's the heater's BOM. Also needs verification. The original had quantity issues.
-
Hi,
Sure that these are the complete BOMs ?
-
Pretty sure. 😉
As I said, they need verification. I did substitute some parts that were no longer available.
-
Then I´ve done something wrong..I count 13 lines on the first excel list(I don´t have Excel on my notebook, so I went to my PC and printed it out as a pdf, maybe a failure occured then)
-
I don't think it was you. This forum (SMF) is buggy sometimes, and it screwed up the files. The main BOM has 44 lines, and I downloaded it, and it showed the heater BOM instead.
Here's both again. Hopefully it works this time. They both need verification for quantity and part accuracy. As mentioned previously, reference numbers were not all correct in the original BOMs.
-
This forum (SMF) is buggy sometimes, and it screwed up the files.
Yup.
Never use the first Choose File option, never !
Instead select more attachments and link yours in there however then you are restricted to 9/post but it sidesteps the forum problems.
-
This forum (SMF) is buggy sometimes, and it screwed up the files.
Yup.
Never use the first Choose File option, never !
Instead select more attachments and link yours in there however then you are restricted to 9/post but it sidesteps the forum problems.
I usually drag and drop, and it usually works. AFAIK, this was the first time it actually confused the files like that.
-
I had/have the same with pics...
When I´ll upload more than one and then in an order, I always ignore the first upload and start with the second.
The DAC in this project..
There´s a "combo" with it and the LTZ1000 reference, together you "gain" a reference with 1ppm accuracy..Nice.
-
Do you think this project is worth the $400+ USD price tag?
-
Honestly?
When you stretch it for month so it doesn´t hurt you directly in case of the costs, why not.
But if you take the money instantly in your hands I would go for the complete assembled, tested and burned-in reference I´ve linked a few posts before.
It do not have the current source, but it is "calibrated" with a 8.5 digit meter.
Me I would go the way of "stretching" this project.
Like I said before, theoretically it could be "tuned" to a 1ppm reference..awesome.
But most of what I expect from it is to learn how things work in detail and then possibly improve them.
That's actually how I've always done it in the past.
-
Well, I ordered the PCBs. 🤦
-
Well, mine arrived today..
-
Here's both again. Hopefully it works this time. They both need verification for quantity and part accuracy. As mentioned previously, reference numbers were not all correct in the original BOMs.
Tested it a few minutes ago by importing the main list into my mouser account.
Batteypack won´t be avaible for me(my region), display won´t be avaible (only for OEM)...
This was the first test, tomorrow I´ll check your list completely.
Display I must get from somewhere else, the housing I like to have it in silver.
-
Here's both again. Hopefully it works this time. They both need verification for quantity and part accuracy. As mentioned previously, reference numbers were not all correct in the original BOMs.
Tested it a few minutes ago by importing the main list into my mouser account.
Batteypack won´t be avaible for me(my region), display won´t be avaible (only for OEM)...
This was the first test, tomorrow I´ll check your list completely.
Display I must get from somewhere else, the housing I like to have it in silver.
Here's the LCD direct from the manufacturer: https://4dsystems.com.au/products/ulcd-35dt/ (https://4dsystems.com.au/products/ulcd-35dt/)
For the regular enclosure, just remove the BK color from the model name: https://www.mouser.com/ProductDetail/546-1455J1202 (https://www.mouser.com/ProductDetail/546-1455J1202)
The battery is uncertain based on size anyway. You might be able to get the original battery from the original BOM (wasn't available to me here in the US).
-
Damn....I think I´ll order the parts next week....
-
Damn....I think I´ll order the parts next week....
Me too. I had to stop myself from ordering them a couple times already lol. But I think I'll have my PCBs on Monday. We'll see how patient I am.
-
We'll see how patient I am.
I felt the same way.
First wait, but then the boards were there and now...crap. ;)
-
Made the order by mouser.
Without housing (can get it cheaper), display, battery, 4mm jacks, several ICs (not in stock) and one resistor (not in stock).
All in 156€...
Hint : The AD5791 will be avaible soon (August) by Digikey
Edit:
Missing parts:
MCP1501T-33E/CHY (11/23 in stock (Digikey) )
MAX6350CSA+ (CSA: 1ppm tolerance) (09/23 in stock (Digikey)
AD8676 (11/23 in stock (Digikey) )
AD5791 (08/23 in stock (Digikey))
Vishay 100Ohm 0.02% 0.2ppm (in stock by digikey)
-
Oh yeah, I forgot to add that $17 resistor. I was planning to find a less expensive but same spec resistor, but I don't think that's gonna happen. 🙄 Oh well. I added Y1625100R000Q9R to the project.
The rest is on the list, but backordered. I have a separate project for the backordered stuff to keep ordering simple. 😉
-
So, display, housing and banana jacks ordered, that´s enough for now and this month.
Forget to order the heater-pcb...Will do it plus ordering the 3D print parts.
Then I´ll have a pause because of the missing semiconductors.
Good.. ;)
-
Well, mine arrived today...
-
Ah, in fancy brown style..
-
It's actually black, that's the lighting playing tricks. Now that I've got it, I paid more attention, and I'm not happy about the micro USB jack. I want to do it with USB-C. I know it's just for charging, but I hate paying attention to micro USB cables.
I'll probably see if I can bodge a charging only USB-C port to it. I think I have enough PCBs to experiment. 🤷
Edit: Who am I kidding, I'll just order a new board version with a USB-C charging port for $10 when I order other stuff. 😉
-
Forget to order the heater-pcb...Will do it plus ordering the 3D print parts.
If you don't rush, I can send you the gerbers when I'm done changing it to USB-C power jack, and seeing if there's anything else that needs attention. There were a couple warnings when updating the PCB from schematic.
I'm glad I saw that line about 3D printed parts, I totally missed that. Oops.
-
Hi,
The housing has already arrived today, looks well made, nice.
BUT:
The board does not fit, it is a tad too wide, maybe 0.8mm ...
This is just to warn you (and others) about reordering the boards.
Sure, you could go ahead and grind one side of the board, but I think that's a bit uncool...
-
That's a minor issue. There are some small variances in the posted dimensions between the different colors of those enclosures. I haven't looked closer than that.
I won't change the PCB dimensions right now because the alignment to the screen is more important than the chassis dimensions, and you either need to trim both sides the full variance amount, or figure out where the dimensional change needs to be made to stay consistent with the LCD mounting.
Well, maybe trimming 1mm off both ends of the board is okay. I'll look at it, I'm working on it anyway.
BTW- there are BOM discrepancies between what KiCad exported and what was supplied with the kit. I'm making a PCBA BOM now to see what small stuff I can get out of the way, and if it's worth it or not.
-
In the BOM, R16 is 100Ω, but on the PCB R16 is 1kΩ. I read the article page, and the original BOM is wrong, the schematic is correct. Also, the schematic was lazy. They wrote 1k but didn't include that it needs to also be 2ppm/C or better.
-
So R16-R20 must have it..OK, thanks for checking this :-+
-
So R16-R20 must have it..OK, thanks for checking this :-+
Yup. When a set of resistors needs to have a specific temp co, then all of the resistors in that group need to match. Or so I'm told. 😉
-
They told you that right. ;)
OK, then I have to get a 1k resistor of the same kind.
https://www.mouser.de/ProductDetail/TE-Connectivity-Holsworthy/RU73X2A1K0LTDF?qs=CXI0wBT1M2iYnqPwpNKfVg%3D%3D (https://www.mouser.de/ProductDetail/TE-Connectivity-Holsworthy/RU73X2A1K0LTDF?qs=CXI0wBT1M2iYnqPwpNKfVg%3D%3D)
-
They told you that right. ;)
OK, then I have to get a 1k resistor of the same kind.
https://www.mouser.de/ProductDetail/TE-Connectivity-Holsworthy/RU73X2A1K0LTDF?qs=CXI0wBT1M2iYnqPwpNKfVg%3D%3D (https://www.mouser.de/ProductDetail/TE-Connectivity-Holsworthy/RU73X2A1K0LTDF?qs=CXI0wBT1M2iYnqPwpNKfVg%3D%3D)
lol, no, a friend of mine. I'm designing another circuit, and we're using an LTC6090, and he made sure to point that out about the tempco. 😉
I updated and verified my BOM vs. the schematic, and yes, that's the correct resistor. Nothing should be missing, but there might be something extra. 🤷
-
Btw, here the evaluation board with LTZ1000 reference:
https://www.digikey.de/de/products/detail/analog-devices-inc/EV-LTZ1000-REFZ/8598410 (https://www.digikey.de/de/products/detail/analog-devices-inc/EV-LTZ1000-REFZ/8598410)
And here the very interesting User guide :
https://www.analog.com/media/en/technical-documentation/user-guides/eval-ad5791sdz-ug-1152.pdf (https://www.analog.com/media/en/technical-documentation/user-guides/eval-ad5791sdz-ug-1152.pdf)
It is already quite similar. 8)
But maybe you can hardly do it any other way....
-
I updated my Mouser project again. AD8676 was the wrong size. All of them are backordered, so you probably didn't order it yet anyway. 😉
-
AD8676 was the wrong size.
I have already fallen for it in 2 other projects, the components arrive - Mehhh...wrong design, does not fit by any stretch of the imagination, reorder again...Annoying.
-
AD8676 was the wrong size.
I have already fallen for it in 2 other projects, the components arrive - Mehhh...wrong design, does not fit by any stretch of the imagination, reorder again...Annoying.
I know the feeling. I did warn you about parts needing to be verified though. I have JLC part numbers for almost everything now too. Unfortunately, most of it is only on their extended parts list. I might see about preordering stuff so they can assemble what's available.
-
My Mouser order should arrive on thursday, friday at work (FedEx is horrible when you´re not at home, so I leave the company adress instead).
Funny thing: Although there´s a Mouser Adress in germany, the ordered goods are coming from the States.
Display should arrive tomorrow from TME.
-
I was also wondering why the display costs so much money...3.5" and touch is not a big burner now.
After I had downloaded the data sheet, however, it was more comprehensible.
Display, input interface and controller for the rest...Well, then it makes sense to me.
-
Basically we're stuck with the screen he designed this with. There are similar screens with similar feature sets for less cash.
BTW- found another error on the BOM. G6K-2 had a Y in it, which changes the spacing. I corrected it in the Mouser project. It's probably good now. 🤦
-
Now the pcb fits into the housing.. ;)
But some parts are coming very close to the housing, must take care when soldering the parts.
And the super-duper-expensive-display has arrived. ;)
OK, the building quality is superb, no question.
First I thought "and how to connect this ?!" - But there´s an adaptor in the package...Puhh. 8)
Edit: I gues for programming the display.
-
It looks like you bought the "gen 4" version of the screen.
You need this one: https://4dsystems.com.au/products/ulcd-35dt/
But it looks like you got this one: https://4dsystems.com.au/products/gen4-ulcd-35dt/
I'm going to reorder the PCB with USB-C on it. I removed .5mm from the top of the PCB, and 1mm from the bottom of the PCB.
-
Brilliant, just brillant.... :palm:
I hate resending
-
Brilliant, just brillant.... :palm:
I hate resending
Did they send you the wrong one? Either way, I'd be surprised if they weren't nice about it.
-
I did a mistake and searched for it by using google.
They also having the right one(30 bucks more)..fu**
-
https://www.tme.eu/de/details/ulcd-35dt/intelligente-displays/4d-systems/ (https://www.tme.eu/de/details/ulcd-35dt/intelligente-displays/4d-systems/)
The right one, on stock...
https://www.tme.eu/de/details/gen4-ulcd-35d/intelligente-displays/4d-systems/ (https://www.tme.eu/de/details/gen4-ulcd-35d/intelligente-displays/4d-systems/)
Mine... ;)
OK, return to sender.
Edit: Pic from his project
-
They also having the right one(30 bucks more)..fu**
And you thought it was expensive before. 😉
-
Oh yes...
At least the program seems to work.
Software: Workshop4 IDE (https://4dsystems.com.au/software/)
And we need a programming cable..
-
Interesting. What's the interface for programming? Direct to chip?
-
I think, it´s this one here:
https://4dsystems.com.au/products/4d-upa/
Or this:
https://de.rs-online.com/web/p/entwicklungstool-zubehor/8417875
-
The first one you listed is for gen4 panels, which we don't want.
The second one might work (but it probably needs a chip; no chip, probably no work). Looking closely at the image of the (correct) panel, it shows RES GND RX TX +5V. That sounds to me like a serial to USB 2.0 converter module is what's used.
Anyway, this is probably the correct one from them (and just looks like a USB adapter): https://4dsystems.com.au/products/uusb-pa5-ii/
That has an FTDI FT232 USB chip on it, so my guess is, if you have another serial USB or Serial over Bluetooth module, you could probably get that to work.
I know this is a crazy question, but did you RTFM for the screen yet? Maybe there's some insight in there.
Did the guy who made the project use the programmer, or was the whole thing done through the microSD card?
-
Hi,
That sounds to me like a serial to USB 2.0 converter module is what's used.
Their programming cable have a serial-usb converter in the connector housing.
I know this is a crazy question, but did you RTFM for the screen yet?
As ever, I flew over the pages... ;)
Did the guy who made the project use the programmer, or was the whole thing done through the microSD card?
The GUI is made as follows:
Touch buttons and digits are images stored on the micro SD card. Each image has x,y coordinates on the touchscreen.
If it´s possible I want to avoid this and program the controller directly.
When I´ll get the right display I´ll take more care to read the manual...
-
The GUI is made as follows:
Touch buttons and digits are images stored on the micro SD card. Each image has x,y coordinates on the touchscreen.
If it´s possible I want to avoid this and program the controller directly.
When I´ll get the right display I´ll take more care to read the manual...
I would be surprised if it's possible. Whether programming directly is necessary or not, I don't know. I didn't read anything yet. 😉
However, I know there are images and sound files involved, so storage is necessary. I don't know how much storage the LCD has without the microSD.
-
That´s a good point.
https://www.elektormagazine.de/labs/voltage-current-calibrator-0-to-10v-and-0-to-40ma-0001# (https://www.elektormagazine.de/labs/voltage-current-calibrator-0-to-10v-and-0-to-40ma-0001#)
Interesting points on the section improvements..
-
Hi,
However, I know there are images and sound files involved, so storage is necessary.
In the datasheet it is written that it is indeed exactly for this, you can also store logger data on it:
The uSD card is used when displaying images/video/sound, along with datalogging to uSD, and a programming cable is definitely required for downloading compiled code and PmmC/Firmware updates.
What is also interesting, so for a programming layman like me:
ViSi-Genie is a breakthrough in the way 4D Systems’ graphic display modules are programmed. It is an environment like no other, a code-less programming environment that provides the user with a rapid visual experience, enabling a simple GUI application to be ‘written’ from scratch in literally seconds.
ViSi-Genie does all the background coding, no 4DGL to learn, it does it all for you.
Pick and choose the relevant objects to place on the display, much like the ViSi Environment yet without having to write a single line of code. Each object has parameters which can be set, and configurable events to animate and drive other objects or communicate with external devices.
Simply place an object on the screen, position and size it to suit, set the parameters such as colour, range, text, and finally select the event you wish the object to be associated with, it is that simple.
In seconds you can transform a blank display into a fully animated GUI with moving sliders, animated press and release buttons, and much more. All without writing a single line of code!
;)
-
I think it will be pretty cool if we can easily customize the interface. However, he already made graphics for it, so it seems like the programming was done already, and it just needs to be copied over? I guess we'll find out eventually.
I'd still bet a serial to USB cable or adapter is all that's needed for the programming cable. I have real programmers too, but I don't know if they'd work with that or not.
-
I'd still bet a serial to USB cable or adapter is all that's needed for the programming cable.
It is :-+ :
https://docs.rs-online.com/d92d/0900766b815118a7.pdf
However, he already made graphics for it, so it seems like the programming was done already, and it just needs to be copied over?
All we need is already there. 8)
-
That's what I thought. So it seems the extent of "programming" is just copying his files over. I may have a look at them with that utility and see what kind of customization can be done.
-
I´ll check this in the next week I guess - Ordered the "right" display today and somewhere I must have a USB to serial adaptor lying around.
-
Stop - Hammertime ! Ehhh...Mousertime. 8)
-
Comparing the BOM (original) with the schematic....
R21(10R) is missed in the BOM but in the schematic avaible and must be there because it´s the output resistor.
Then it´s getting "wilder"...
2µ2 tantalum are only mentioned as C9,C10 in the BOM but also avaible as C30,C33,C36 in the schematic.. :P
According to the datsheet of the MAX6350 (U9) C30 and C36 are right with 2µ2 (and wrong in the BOM), for C33 (100nF in the BOM) the datasheet recommending 1µF, in the schematic it´s 2µ2.
And another thing:
C28 is 10µF in the schematic (and right), but 100nF in the BOM....
I´ll continue checking BOM vs schematic.
-
Your BOM must be outdated. I have a new BOM I verified with the schematic. Try the attached BOM instead. I don't remember if I updated my project yet. I got sidetracked with...work. 😉
-
Argh, excel again... ;)
So you also found out that D4 does not exist?
I looked at the MAX6350 datasheet again, 1µF for the NR input is recommended, but if it's more it won't have any effect, so the 2µ2 can stay that way.
-
I exported a BOM from Kicad and verified that way. Much easier than the original goofy BOM that had some bad info in it.
-
The right display arrived... :D
But it seems not so easy just to take a common usb-serial adaptor and then everything´s fine.
I got a notebook with a native com port, tried this first...nothing.
Then took the adaptor, tried again...nothing.
The display won´t start when the RX/TX pins are connected to the com port or to the adaptor.
Maybe the RESET pin plays a role too, must examine it further.
Or buy the original adaptor. 8)
(adaptor pic made before changing the connection)
-
Your pictures don't make the wiring clear. Do this:
If you're powering from an external PSU, connect power and ground to the screen, and make sure ground also goes to the serial port adapter.
Connect Rx/Tx backwards at the serial port adapter and see if that works.
-
Thank you josh, but this was already made.
Will continue testing.
-
Thank you josh, but this was already made.
Will continue testing.
Do you mean you got it working?
-
Sorry, the connections were meant...
-
I have now rather stopped experimenting, not that I still break the 120 euro display. ;)
USB to serial TTL converters are available like sand on the beach and super cheap.
And many look almost exactly like the "original" from 4D.
I ordered one from Amazon and it will arrive on Monday.
It's so cheap that it doesn't hurt if it doesn't work.
Because the adapter from 4D has something that all the others don't have, the reset output.
And it is also triggered, for a hundred milliseconds.
If it turns out on Monday that the ordered adapter does not work with it, well then I have to buy the original. 8)
I have made again parts list control, what do I have here, what am I still missing....
I soldered nothing yet, I will take the board with me to work and let the coils L1/L2 and the USB port soldering there because no way to do it here.
Oh, and the Heater board is now also ordered.
-
I usually use the FT232 USB chip stuff, I'm guessing that will work when I eventually get this. I'm also curious if my PicKits would work or not.
If you know which pin to connect the RES to, then it may work with what you have. I think RTS (pin 7) is the correct pin.
-
If I remember it right it is DTS, will confirm this yes/no later.
What I haven't quite figured out yet is why the display turns off/not even starts up once the RX/TX lines are/will be connected.
And I know about potentials, believe me.... ;)
-
I looked at the MAX6350 datasheet again, 1µF for the NR input is recommended, but if it's more it won't have any effect, so the 2µ2 can stay that way.
The NR pin is rather high-ohmic and has direct influence on the output voltage stability.
I would never use a tantalum (with high leakage current) here.
I would use a film capacitor. (or leave it unpopulated since it only helps with wideband noise).
with best regards
Andreas
-
Most of the cheap USB to uart adapter are just that: UART output (3.3 or 5 V level), similar to what a µC provides or wants. So there are no drivers to get the +-10 V levels for RS232 like a generic PC serial part.
+1 with using a film capacitor for the NR pin of the reference. A tantalum or Al electrolytic is more a no go. A ceramic X7R or similar may work, but is not ideal. Even if only for the higher frequency noise, it helps to have at least some filtering. 1 µF film caps (MKS) are still afordable and not excessively large.
-
Where are the UPDATED Kicad files for this project? And where is the link to the ORIGINAL Elektor article? Thanks.
David.
-
MKS caps are only radial, it needs to be SMD for this project. 25MU105MA23216 25V 1µF 1206 SMD 20% would probably be a good 1µF film substitute to fit the same footprint. Not cheap, but certainly not that expensive compared to other parts in this project.
Why wouldn't an X7R cap be good for that? C1206C105J3RACAUTO 25V 1uF X7R 1206 5% is only 40 cents instead of $4.00.
-
Hi Josh,
MKS02 wouldn´t be a problem, I just test it out...
-
Where are the UPDATED Kicad files for this project? And where is the link to the ORIGINAL Elektor article? Thanks.
David.
Link see first post.
Updated Kicad files ask Josh... 8)
-
Hi Josh,
MKS02 wouldn´t be a problem, I just test it out...
There's a plastic enclosure that covers that area for the heater block. MKS won't likely be a good idea/fit there. You can try when you get the 3D printed stuff in and see how tight it is.
Sorry, I'm not sharing my kicad files. I blew a lot of time on them, and I refuse liability for any errors anyway. 😉
-
You can try when you get the 3D printed stuff in and see how tight it is.
Or make it a litle bit deeper.
Sorry, I'm not sharing my kicad files.
Eehhh... ;)
-
Pic showed wrong place (C30), now on the right place.
-
So now it's going to be compressed between U9, and the 3D part that won't fit if you don't change it. Not worth it. If you're going to spend $500 on parts to make a VA reference, spend the $4 or 40 cents to get the appropriate part. The entire design is only as strong as the weakest point you introduce.
-
Well,
As long as not all ICs are avaible yet, we´ll have all the time to think about it.
BTW, the AD8676ARZ we need is avaible in Ebay, from a chinese vendor....
Fake or not.. (https://www.ebay.de/itm/394754636287?hash=item5be935b9ff:g:HO8AAOSwR-tkpczt&amdata=enc%3AAQAIAAAA4NwBGWEWxyvX%2FuU9DOauoJ4LMo36TqStADg8AWw5QctjLQxLNZHktHbK15pZtmTXS8dWRke6Ms%2Bc0sYu%2BV8%2FLyP8qyjSGu2cZ7hWcCX5bhZCTZRzxbploxi7vW4tL2A5sgXmHr7MmwmoyS%2BQIuNMDSlyN6etAFBBSTnPenhuRGgbaMrn9ofgQtNEaUuYMOwQhGS0YF8J%2BsIWH6aBkOr3w7ul%2FxrQN0rN85vATdJWob1evosLqc0F%2FYwsEosJGpQft0G86jLJ3dk7kSRNCGSARi6PifyNxNmtRGolpRPeQxx3%7Ctkp%3ABk9SR5K-wYG1Yg)
-
I'd be patient, same statement as above regarding introducing weaknesses into your expensive project.
I also found somebody who would sell the expensive chip for $70, but I don't trust them. 🤷
-
Yes, I also think I will first equip everything I already have and then we'll see.
As mentioned, tomorrow the USB port and the chokes soldered, then the thing with the display and if that has all worked, you can continue with the housing and eventually the missing ICs are available again. ;)
-
Well,
As long as not all ICs are avaible yet, we´ll have all the time to think about it.
BTW, the AD8676ARZ we need is avaible in Ebay, from a chinese vendor....
Fake or not.. (https://www.ebay.de/itm/394754636287?hash=item5be935b9ff:g:HO8AAOSwR-tkpczt&amdata=enc%3AAQAIAAAA4NwBGWEWxyvX%2FuU9DOauoJ4LMo36TqStADg8AWw5QctjLQxLNZHktHbK15pZtmTXS8dWRke6Ms%2Bc0sYu%2BV8%2FLyP8qyjSGu2cZ7hWcCX5bhZCTZRzxbploxi7vW4tL2A5sgXmHr7MmwmoyS%2BQIuNMDSlyN6etAFBBSTnPenhuRGgbaMrn9ofgQtNEaUuYMOwQhGS0YF8J%2BsIWH6aBkOr3w7ul%2FxrQN0rN85vATdJWob1evosLqc0F%2FYwsEosJGpQft0G86jLJ3dk7kSRNCGSARi6PifyNxNmtRGolpRPeQxx3%7Ctkp%3ABk9SR5K-wYG1Yg)
The AD8676 is not that unique. There is a very good chance that one could substiture with a similar part that is better awailable, like AD8672 or ISL28227 or depening maybe also slower chips.
-
USB port and the coils, soldered at work...
Looks good ( bad light for the pics) and I could start soldering the avaible components.
Maybe next weekend, until then a part of the Digikey order should have arrived, with the sinfully expensive 0.01% 0.2PPM resistor in 1206 format, which alone costs almost 15€, I ordered then also only one... ;)
Heater pcbs are shipped.
-
I couldn't tell if you got the programming cable situation straightened out. If not, 4dsystems has detailed schematics for the uUSB-PA5 which is a small PCB version of the programming cable.
- https://resources.4dsystems.com.au/datasheets/accessories/uUSB-PA5/
The schematic shows that the RESET pin of the programming cable is controlled by DTR through a transistor. If DTR is asserted (5V) then the RESET pin is routed to GND. So I don't think you'll be able to just use an off-the-shelf USB-to-TTL serial cable. You certainly can't use a serial port that's using RS232 voltage levels.
-
Today the digikey order arrived, except the MAX6350 and AD5791 "of course".
But they will be delivered later, if in stock - So I can not forget about them.
I don't have to worry about the two AD8676 anymore, thanks to a very nice user. :)
So today the missing resistors and capacitors arrived.
C33 is now a foil with 1µF.
The resistors look quite unspectacular, for that they were so expensive. ;)
The 100 Ohm (R10)with 0.02% tolerance and 0.2PPM Tempco has very narrow solder joints, the 1K (R16) in 0805 case and with 0.01% tolerance looks like any other.
And I'll take the board to work again tomorrow, there is one more component after the L1/L2 and USB port that I can't solder at home:
U3... :P (because of the exposed metal back)
Hopefully it will work tomorrow, I wanted to solder at home this weekend what I already have.
-
Today the digikey order arrived, except the MAX6350 and AD5791 "of course".
AD586L or MAX6250A play in the same ball park as MAX6350 (especially if the heater option is populated).
After my experience the AD586 has better long term stability than the MAX references. (at least in DIP8 package).
with best regards
Andreas
-
Thanks for the tips and hints, I will keep them in mind.
The current project is well advanced, I have all the components except the reference and the dac IC, I have the board with appropriate layout, I will finish accordingly and see how it performs.
-
And I'll take the board to work again tomorrow, there is one more component after the L1/L2 and USB port that I can't solder at home
You don't have hot air or an SMD PCB heater?
-
No Sir.. ;)
At work we have reflow, infrared, damp heat, etc., so I don´t need cheap solutions at home which could damage the part easily.
-
It's not usually the solution that damages the part. 😉
-
With the cheap hot air stations or these homemade reflow ovens, it's simply a gamble and not a solution and as long as I have access to proper tools, I'll use that. ;)
The colleague had already soldered the chip today, so I could take the board home again.
Earlier I have first marked the positions, pin 1 of the ICs, anode of the diodes, plus of capacitors ... ::)
Today also the USB to serial converter arrived(a second, the first order was getting "lost"..)
Would have been nice, there would be a small drawing of what exactly you can do with the small switches... :P
-
Would have been nice, there would be a small drawing of what exactly you can do with the small switches... :P
I found these diagrams which should get you going:
-
Today I wanted to solder all ICs that I already have here, that worked so far well until I opened the bag where U7 is packed inside:
https://www.youtube.com/watch?v=oQ90FE5BfvA (https://www.youtube.com/watch?v=oQ90FE5BfvA)
-
Today I wanted to solder all ICs that I already have here, that worked so far well until I opened the bag where U7 is packed inside
Hmmm, that sounds familiar... 😉
I have already fallen for it in 2 other projects, the components arrive - Mehhh...wrong design, does not fit by any stretch of the imagination, reorder again...Annoying.
I know the feeling. I did warn you about parts needing to be verified though. I have JLC part numbers for almost everything now too. Unfortunately, most of it is only on their extended parts list. I might see about preordering stuff so they can assemble what's available.
-
Today I wanted to solder all ICs that I already have here, that worked so far well until I opened the bag where U7 is packed inside:
https://www.youtube.com/watch?v=oQ90FE5BfvA (https://www.youtube.com/watch?v=oQ90FE5BfvA)
If that's SOIC vs TSSOP you might be able to bodge it in an emergency. If pinout is the same. Bend up corner pins. Solder the center four pins. Then four small wires for the corner pins. I've done something like that once.
-
Hi,
I had already thought about that and I will do the same for another project (curvetracer).
But not for this one, especially since the chip is abundant, even in the SOIC housing.
But I am currently collecting something that I can still order, just the IC alone hardly makes any sense. 8)
-
You can modify the PCB to accept both IC packages on another side seems to be place. You only need to order new PCB ;D.
-
This appears to be the correct package: https://www.mouser.com/ProductDetail/584-AD8276BRZ (https://www.mouser.com/ProductDetail/584-AD8276BRZ)
I'm gonna look over those chips and update the BOM. I already updated my project.
-
AD5791 BOM says SSOP, all versions in reality say TSSOP.
PZT2222AT1G is the model specified, but mouser says SOT-23-4, and shows SOT-23-3 in the photo (3 pin is correct). 🤷
I updated the BOM again (attached).
One other part number was off, but I don't know if that matters or not. It said 2N7002-T1-E3 instead of 2N7002E-T1-E3.
As I said before, this is still not fully verified. 😉
-
Heater pcb finished - And it works.
Initial (and max) current appx 257mA, then falling, rises again when the pcb is getting "colder", so the regulator circuit is ok.
-
I have now tried the adapter - With zero success.
Does a RESET signal really have to be present, or have I already destroyed the module before, that is the question.
What is good:
The adapter has the DTR signal out as a pin, so you could put the reset circuit in between.
For this you would have to know what values the RC part has and if the transistor is a PNP or NPN type.
It's a pain in the ass to spend almost 40€ for the original programming adapter, but if I don't get further tomorrow, I will order it.
And if it works then, examine the adapter more exactly and make it public, so that others do not have to spend the money. ;)
-
I used the Waveshare FT232 TTL module and connected it to the original programming adapter as follows. The green wire is the DTR signal. I set the voltage level to 5V. With this it worked for me.
-
I've only just found this thread. I built the Elektor Voltage Current Calibrator about half a year ago.
If the original circuit boards are used, the components on one side are too close to the Hammond housing and can cause a short circuit.
-
Hi Tom,
Glad to hear that you already did it :-+
Isolation:
Yepp this is known, I´ll isolate the edges with capton foil when assembling the board into the housing*.
Display:
I made it... 8)
I wondered why the RX/TX leds on the adapter never blinked when the program scanned the port.
So I connected my scope to the channels and lo and behold, nothing...Nothing at all.
Either the adapter is defective or the configuration was not right yet.
It was the latter....the DIP switch must be set as shown in the picture, then it works.
(thanks again to mwb1100 for the diagrams (https://www.eevblog.com/forum/testgear/elektor-voltage-current-calibrator-project/msg4996720/#msg4996720), it must the first configuration.)
So you don't have to buy an expensive adapter from 4D systems...Very good.
Now I have to buy a micro SD card. 8)
*) If everything goes well afterwards, I will probably take another housing.
-
Mouser now have the MAX6350CSA on stock again, order it and the U7 in the right package.. ;)
So the AD5791 DAC is the only thing missed, will be in stock in september, october.
But like the other ICs from AD, there are at least two types, A-grade and B-grade.
Where B-grade have the "better" specs in some cases.
-
Now I have to buy a micro SD card.
Arrived today, I then formatted it to FAT16.
What I then need to figure out is which files need to be saved to the SD card and which (or which) file is intended for programming the diablo processor.
At the moment I think that the content of the USD_files folder belongs on the card...
I will try the days, what belongs to what.
-
The Software folder with the En_course subfolder is the software that would get programmed with the WORKSHOP4 4D Systems programmer software.
The USD_files folder is what goes on the microSD card.
-
Sounds logical, we'll see.
-
My mouser order arrived today at work, now the DAC is the only thing missed.
And U7 have now the right size...Soldering it and the MAX6350CSA+ in a break, like the tantalum capacitors and the one film cap.
We have also a 3d printer here, so chances are good that I could make the frontplate and the housing for the heater at work.
Next weekend I´ll try to program the display.
-
Next weekend I´ll try to program the display.
Couldn´t wait, did it today in the evening.. ;)
Success.. 8)
Tried several times, several things, at the end it was easy:
- You musn´t open the 4D Workshop main program.
- In windows start, click on the 4D mainfolder and start "Program Loader" (connect the display before via usb adapter to the pc/notebook)
- Choose the right com port.
- Select the Software ->En_cours folder, the executable file will be displayed.
- Click to OK to start programming
- That was it for the cpu part.
SD-Card:
- Must be formatted in FAT16, I´ve bought extra a 4GB card for it and formatted it via windows command line.
- The 4D formatting tool (you can format cards above 4GB with it) won´t work here.
- Copy the files from the USD folder directly to the card
I did this after the programming, before, while the first start of the display, the card was empty and you see only a "+" symbol and a dot on it.
After all I´ll test the display and tap the +/- fields - It works...
So this is finished, next monday I´ll solder the rest on the pcb (loose one 10µF tantalum, must look here at home if I got another).
Martin
-
Nice. That's pretty typical regarding the programmer. Programmers and Development environments are often bundled together, but actually separate.
-
Oh yes, I´ve studied the manual before, but didn´t find a thing about it.
BTW, another reason why the display is so expensive:
Not only it is "intelligent" (controller for other things), the display itself got the highest selection grade.
Only 0...4 max. dead pixels.
-
Finally in stock:
https://www.mouser.de/ProductDetail/Analog-Devices/AD5791BRUZ?qs=NmRFExCfTkEaSRRmY4%2Fg1g%3D%3D (https://www.mouser.de/ProductDetail/Analog-Devices/AD5791BRUZ?qs=NmRFExCfTkEaSRRmY4%2Fg1g%3D%3D)
And ordered, should arrive next week - The last one I need for completing the circuit... 8)
-
Did you build this yet? I'm curious how it came out.
-
Hi,
Unfortunately not yet.
The chip has arrived and is together with the board on the table at work.
I wanted to supply the board with voltage to see if everything is ok before I solder the most expensive part of the circuit. 8)
Too much to do the last few days.
I think, soon it will be so far. :)
-
I wanted to supply the board with voltage to see if everything is ok before I solder the most expensive part of the circuit.
And I already know how, I haven't soldered in the PS1 converter yet, so I can connect two power supplies with current limiting to the +/-15V inputs and carefully boot up....
I am quite sure that I have observed the polarities of all components, but "thanks" to the missing markings on the board, something may still have gone wrong, hence my cautious approach.
-
Did it today..
No anormalies....Wait a minute..
Why do I measure 2.5V on the output of the MAX6350 ???
Because it´s a 6250... :P
Overlooked this, now I´ve ordered the right one - and another part which is wrong:
PS1....
In my "old" BOM its a single 15V, we need a dual 15V... |O
Mouser-Nr.:490-PEME2-S5-D15-S is the right one(if you don´t have your BOM updated @Josh)
Martin
-
Did it today..
No anormalies....Wait a minute..
Why do I measure 2.5V on the output of the MAX6350 ???
Because it´s a 6250... :P
Overlooked this, now I´ve ordered the right one - and another part which is wrong:
PS1....
In my "old" BOM its a single 15V, we need a dual 15V... |O
Mouser-Nr.:490-PEME2-S5-D15-S is the right one(if you don´t have your BOM updated @Josh)
Martin
You didn't order 700-MAX6350CSA for the 6350? That's what's on the BOM.
If the single output is wrong, then the original downloaded BOM is wrong too. The BOM that comes with the kit says "IH0515SH (https://www.mouser.com/ProductDetail/XP-Power/IH0515SH?qs=w%2Fv1CP2dgqr5cWZwR0OP4g%3D%3D)" - which is single output. Are you sure about that part change? Did you also verify the footprint?
Thanks,
Josh
-
I know now what went wrong...
The MAX I´ve googled instead of using the BOM, the link to mouser brings you to the 6250 - Although in my search I´ve typed 6350... ::)
Thank god the 6350CSA+ is avaible.
(There is also a 6350CSA+T, with 0.5ppm/C° avaible, but with a smaller package.)
Are you sure about that part change? Did you also verify the footprint?
Absolutely.
Look at my post, the pic:
https://www.eevblog.com/forum/testgear/elektor-voltage-current-calibrator-project/msg5094678/#msg5094678 (https://www.eevblog.com/forum/testgear/elektor-voltage-current-calibrator-project/msg5094678/#msg5094678)
Plus the op-amps on this board are needing a dual-supply(output of the reference is +/-10V)
Martin
-
Oh well, the original BOM screwed us again. Well, mostly you since I didn't order shit yet. 😉
-
If you wait a little longer to run my reference, we'll have all the bugs worked out.
My loss, your gain. ;)
-
If you wait a little longer to run my reference, we'll have all the bugs worked out.
My loss, your gain. ;)
Martin the guinea pig. :-DD
Having trouble with that image now......
-
;D
Someone has to do it...
-
If you wait a little longer to run my reference, we'll have all the bugs worked out.
My loss, your gain. ;)
That was always the plan. 😉😉🤣🤣🤣
-
Why do I measure 2.5V on the output of the MAX6350 ???
Because it´s a 6250... :P
Hello,
Shure?
the MAX6250 is also a 5V reference. (with 2 ppm/K)
The 2.5V versions are MAX6325 or MAX6225(A).
I would check the supply voltage of the reference on Pin 2. (Minimum 8V necessary).
with best regards
Andreas
-
It works !!
Soldered up the DAC at work today, then took it home and plugged in the display and runs right out of the box.
+10V/-10V, 0-40mA, no problem...Very good. :D
Quick and dirty set up, took a few pictures.
Soon there will be nicer pictures.... 8)
But the main thing is that it works already. :)
@Andreas:
Yes, that was completely the wrong one, I was wrong, it was a 6225.... :P
-
Awesome!
So it seems to be accurate only within 3 digits, or your meters are. I'll assume the meters are more accurate. Are you going to have it calibrated? Does it have full adjustment settings?
-
Hi,
I just put it into operation for the first time, without any other actions.
These will follow when the board, heater and display have been moved into the case.
I first wanted to know if it works at all....
-
Hard to compete with Ian's reference.
On Mouser using BOM from Josh the price is 366.67 E. And this is without hammond box, and pcb (which I have them ordered about two years ago when I try to start building the project but stoped because of the lack of DAC on market).
At about the same price you get a ready calibrated and very stable source with PDVS2 mini.
Also this project need calibration with a metrology grade tool.
Just DIY spirit justify spending so much for this project.
-
Just DIY spirit justify spending so much for this project.
Exactly. ;) And I still have fun doing it.
Apropos hammond housing:
I must have got a copy, which is rather too narrow in terms of dimensions, because a few components threaten to collide with the grooves ... I'll have to think of something else.
-
...
Exactly. ;) And I still have fun doing it.
...
I think this is the spirit :)
When I ordered I used the code from the project (HAMMOND 1455J1201BK) but is the same situation. I think the problem is from pcb design, the author should've leave a clearance between the metal case and the component zone, 2 more mm would be enough.
-
I have taken the information about calibration and the benefits of the heater from the website and put it into a text document.
Edit:
but stoped because of the lack of DAC on market).
The DAC is available again.
It was before, but the "very cheap" version is somewhat "worse" in terms of performance.
The same with the voltage reference, Max6350 is not equal to Max6350.
The CSA+ type is the one with the 1ppm, others are above, one has 0.5PPm (CSA+T).
Then the OP amps, "A" grade is in this case the "worse" variant, I have all in "B" grade.
Then I will soon look for a permanent place for the board and keep it in operation for more than a month...
-
Don't believe my old attachment BOM without verifying first. I'll export a new one eventually.
Yes, it's 100% the spirit of DIY that makes this project desirable. My $20 AD584L voltage reference covers my needs in reality.
-
Hi,
The 3d printing parts I´ve ordered now from here(website only in german)
https://www.3ddesign24.de/ (https://www.3ddesign24.de/)
Well, let's see if they will also grumble because of the partial thinness, like JLPCB. ;)
-
Hm-Hm...
The battery according to BOM I could not get at Mouser, had not interested me at first.
But now... ;)
Searched and found at Buerklin:
https://www.buerklin.com/de/p/mikroelektronika/akkupacks/mikroe-1120/74S4080/ (https://www.buerklin.com/de/p/mikroelektronika/akkupacks/mikroe-1120/74S4080/)
OK, 2-pin connector, that will be connector J2 on the board.
But then why the multi-pin header J1 on the board with the note "Battery Boost Option" ?
A look at the schematic reveals that something else must be connected to the connector, the battery on J2 alone is not enough.
Especially not the voltage, the 3.7V are not enough for PS1.
So an additional circuit is probably connected to J1, in the schematic it will be called "Step up charger"
But there is nothing to read about it in the documentation...
-
Earlier glued the Heater PCB to the designated place on the motherboard, using thermal adhesive.
When it has set, the heater housing is attached.
The two housing parts were also the only thing that was in order from the delivery of the online 3d print provider.
The two front parts I have criticized, my brother with his 200 € printer would have done the same or better. :P
Now better manufactured come.
-
Cured overnight, holds bombproof.
The rest is easy, soldered wires and put the housing on it.
Then switched on again, everything still runs, the area is warm and in the display is also no longer a warning message.
From the wiring diagram, the heater is supplied directly via the usb connection, past the switch.
That is, he heats even if the reference is off, which makes sense.
Next steps are the final installation in the housing and then let age. ;)
-
Did you ever finish this? I might actually make one soon.
-
Hi Josh,
I'm still struggling a bit with the housing.
The 3D parts that were delivered later and manufactured again are just as badly made as the first ones.
Then I wanted to install 4mm safety sockets, but there is only enough space for the normal 4mm banana sockets, which I don't like.
However, I have planned to simply run the module soon because of aging before calibration(at least for a month).
And in the meantime I might buy another, somewhat more spacious housing.
-
I dunno if you looked at my cal chart in the LCR thread, but the two voltage sets are my two different Siglent DMMs. The 3055 is the one that was high, and the X-E was the one that was in spec for the LCR. So I want to get a better reference to see which one is actually closer.
-
I understand that and you won't be able to avoid having this reference calibrated externally, just like me.
Because it has deviations which are in the nature of things.
The manufacturer recommends a multimeter with at least 6.5 digits for calibration (I have one), but in order for me to be able to test my multimeter with it, a calibration with at least 7.5 digits would be necessary.
We have seen that the reference from the first start is certainly not as correct as it could be.
And the aging only serves the purpose of being able to take the later calibration "seriously" in the sense of not having too many fluctuations afterwards.
The benefit for you as a later rebuilder will be how accurately the reference can be calibrated.
If it can keep to the recommended 10ppm, you can start rebuilding it, otherwise I would leave it alone.
So look forward to the next few weeks with me. ;)
-
Yeah, that's also why I'm torn between this project and purchasing an already (actually) calibrated voltage ref. This project would be the more fun option.
Edit: I just ordered the vref10-001 r9. Significantly less expensive, calibrated, and includes 2 years of free calibration. I'll wait on this project until necessity or boredom bring it on. 😉
-
I think my lazy idea to get that 10V ref is crap. I need the versatility this project has, and I still need it calibrated. So I think the first thing I'm going to do is let my 3055 run for a week or two then send that in for calibration and compare my meters. If it's still questionable, then I'll probably build this project. I'd be more inclined if the housing wasn't an issue...though I do have a friend I could bother to make the housing design better.
Right now, my X-E is more stable than the 3055.
-
That is the current status.
The 3D parts and the cover for the display do not fit, the bolts are too thick for the openings (well, you could drill them out), overall the housing provided is very tight.
I am looking for a slightly larger housing.
-
I think my lazy idea to get that 10V ref is crap. I need the versatility this project has, and I still need it calibrated. So I think the first thing I'm going to do is let my 3055 run for a week or two then send that in for calibration and compare my meters. If it's still questionable, then I'll probably build this project. I'd be more inclined if the housing wasn't an issue...though I do have a friend I could bother to make the housing design better.
Right now, my X-E is more stable than the 3055.
With less money than you spend on this project, you can get PDVS2 mini, which is already calibrated and that has stunning stability.
I was already tried to make this project but I have made the correct choice at the right time and I don't regret.
-
I was already tried to make this project but I have made the correct choice at the right time and I don't regret.
We'll know if that wasn't regrettable when I get this reference back from calibration.
But yes, a finished product is of course tempting, especially when it's cheaper.
In addition to the "fun" of building it myself and learning in the process, I also found the possibility of generating a reference current tempting.
Oh, we'll see, of course I hope it was worth all the money. :-X
-
The housing has arrived.
Well, blue is not the color of choice, but it's not that terrible.
I got to work and created a new front panel in the front panel designer, I can upload the file to the store and get it in about a week.
I'm currently thinking about aluminum "nature" or blue... ;)
-
Nice, what brand/model enclosure is that?
-
https://www.amazon.de/dp/B0CB8GXMSB?psc=1&ref=ppx_yo2ov_dt_b_product_details (https://www.amazon.de/dp/B0CB8GXMSB?psc=1&ref=ppx_yo2ov_dt_b_product_details)
You can also get this housing from other suppliers on amazon in different sizes.
For the money, this case is neatly constructed and made mainly of metal.
-
I got to work and created a new front panel in the front panel designer, I can upload the file to the store and get it in about a week.
Hmm, that's a good point. I could make my own panels on my laser. 🤔
-
Was PEME2-S5-D15-S the part you used for PS1? Did it have 5 pins? I'm assuming that if that's the correct part, then the Mouser pictures are wrong?
Thanks,
Josh
-
Hi Josh,
Yepp, it´s the PEME2-S5_D15-S....
Martin
-
Thanks. Mouser seriously needs to fix their images.
-
You said the Hammond enclosure was too small. In which dimension?
This is the original part number from the original BOM:
1455J1201 1455J1202 L:120 W:78 D:27
If it's too narrow, the 1455K1201/1455K1202 increases depth to 43mm. There are other options for increased dimensions: https://www.hammfg.com/electronics/small-case/extruded/1455 (https://www.hammfg.com/electronics/small-case/extruded/1455)
Thanks,
Josh
-
Hi Josh,
The inner width of 75mm +/- 0.5mm is a very "precise fit", as the circuit board also measures 75mm.
You have to stretch the housing a little when inserting the circuit board, but of course it holds firmly.
The main problem is part of the positioning on the board, an area is a little under 3mm from the edge of the board - this can lead to collision problems, as the grooves inside the housing protrude approx. 2.5mm.
The component height of the capacitors in the area was also a problem for me, which I could only solve with a lot of "dremeln".
The 27mm depth, on the other hand, is only a problem if, like me, you want to use the larger 4mm safety sockets, otherwise not.
Martin
-
So it sounds like for the best options, I should reposition those components with proper clearance from the board edge, and switch to 1455K1201/1455K1202 for the extra depth in case I want safety sockets.
Edit: Actually, looking at my board version. I already reduced the size to 73.5mm, but I could easily shift everything down a couple mm. I think I'll slightly increase it to ~74mm, and shift everything for clearance.
-
If the ICs weren't so expensive, I would also put my hand on the design again...
Be that as it may, I have now ordered the front panel in aluminum "natural", the blue was a bit too daring for me in terms of appearance. ;)
The housing isn't huge, but it's big enough to be able to install a stationary power supply - maybe I will.
A little gag on the side:
Not too long ago, I was looking for an electrical load that could provide small currents precisely so that I could compare current clamps in the mA range with it - but I already have a precision current source with deviations in the µA range...LOL. :P 8)
-
I changed the board height to 74mm, and increased the clearance. There's over 4mm clearance from the closest component to the edge. It's hopefully safe now for either enclosure without any "improvements." 😉
If the ICs weren't so expensive, I would also put my hand on the design again...
What else would you change?
-
I would implement mounting holes to be more independent from the housing and several testpins for the different stages for easier debugging.
Further I would think about to "erase" the whole battery circuit as the heater is a real "current sucker".
But first I want to finish this puppy.
Martin
-
I would implement mounting holes to be more independent from the housing and several testpins for the different stages for easier debugging.
Further I would think about to "erase" the whole battery circuit as the heater is a real "current sucker".
But first I want to finish this puppy.
Martin
I haven't thoroughly looked yet, but there might already be some test points. I recall DRC mentioning VIAs only connected on one side, and not having any other purpose. One of those was on a +13.5V line, which I assumed was an intended test point.
I'm probably skipping everything to do with the battery anyway, but I don't see a purpose in removing it from the board in case I ever change my mind about wanting it to be portable for 5 minutes. 😉
-
What test points would you add? I want to see if there are any other changes I should make before ordering the board...again. 😉
-
I would have to check the circuit diagram again, but in principle all voltages generated by the board (3V3, 10.5,V -10.5V, as well as the reference output of the MAX chip (5V).
I think it is easier and safer to measure at such test points instead of measuring directly at the components with test probes.
-
I would have to check the circuit diagram again, but in principle all voltages generated by the board (3V3, 10.5,V -10.5V, as well as the reference output of the MAX chip (5V).
I think it is easier and safer to measure at such test points instead of measuring directly at the components with test probes.
That's what I figured. I know there's that via I mentioned for the 13.5V, I'll look again and see if there are already others hiding.
-
What do you think about this? I added test points for 3.3V, 5V, +/- 10.5V, and +/- 13.5V (wrote 10.5 instead of 10.49 to save space).
(https://www.eevblog.com/forum/testgear/elektor-voltage-current-calibrator-project/?action=dlattach;attach=1993084;image)
-
Nice. :-+
Ref_out is missing, but if I remember the circuit diagram correctly, we can measure this over a cap.
-
Ref_out is missing, but if I remember the circuit diagram correctly, we can measure this over a cap.
I'm confused. What point is Ref_out? Output has its own test point already, and the voltages marked REF are the same as the 10.49V test points.
I added a few test points on the rear (for no reason 😉), and a better place to jumper the VIN + VOUT if not using a battery (ignore the 3D connector, I would jump directly).
(https://www.eevblog.com/forum/testgear/elektor-voltage-current-calibrator-project/?action=dlattach;attach=1993126;image)
-
What point is Ref_out?
I meant the output (pin6) of the reference (U9, Max6350).
But I have just looked at the circuit diagram again, you can measure this voltage at C36, so an extra test point is not necessary.
-
I meant the output (pin6) of the reference (U9, Max6350).
But I have just looked at the circuit diagram again, you can measure this voltage at C36, so an extra test point is not necessary.
Ah, I see. I added one anyway. Thought it won't be so useful once the heater crap is in place.
(https://www.eevblog.com/forum/testgear/elektor-voltage-current-calibrator-project/?action=dlattach;attach=1993162;image)
-
Thought it won't be so useful once the heater crap is in place.
The heater board is on the back, but you're right, it would only be interesting for the first test.
-
Thought it won't be so useful once the heater crap is in place.
The heater board is on the back, but you're right, it would only be interesting for the first test.
Yes, but the 3D printed heater enclosure covers that whole section. As you said, it's only useful for initial testing anyway.
-
My engraved front panel has arrived, in superb quality.
It also fits perfectly in the case, and then there was something else in the package - I don't even know how they came up with it.... 8)
-
What did you use for R9? It's 0Ω in the schematic. I'm probably gonna use RC0805FR-070RL.
-
I would have to look again in the delivery bills to see which type I ordered.
Definitely a Zero Ohm.
-
Good. I found one of the 10k part numbers pasted there by mistake. Dangerous BOMs. Oops. 😉
-
Yes, you can really only rely on the wiring diagram.
I had looked through the delivery bills and then I remembered.
I had forgotten to order it and took an 0R from Vishay from our laboratory stock,
-
What part numbers are you going to use for your external connections?
-
If you mean the connections for the measuring device, these will be 4mm sockets from Hirschmann or MC Contact.
https://www.conrad.de/de/p/sks-hirschmann-seb-2610-f4-8-sicherheits-laborbuchse-buchse-einbau-vertikal-stift-o-4-mm-rot-1-st-734907.html?refresh=true (https://www.conrad.de/de/p/sks-hirschmann-seb-2610-f4-8-sicherheits-laborbuchse-buchse-einbau-vertikal-stift-o-4-mm-rot-1-st-734907.html?refresh=true)
-
I still had 2 sockets, very good.
The 3D parts are still good for something, one frame is perfect as a spacer.
It looks good with the front panel, but I still have some tinkering to do because the reference board has to be mounted on the base of the housing.
It will be... ;)
-
Mouser has those too: https://www.mouser.com/ProductDetail/845-972355101 (https://www.mouser.com/ProductDetail/845-972355101)
Noice.
-
Ah, that's why I didn't find anything when I entered "Hischmann" in the Mouser search function....Same model, different brand.
-
That front panel looks great. :-+
-
Ah, that's why I didn't find anything when I entered "Hischmann" in the Mouser search function....Same model, different brand.
It's actually made by Hirschmann for Altec. It shows both logos if you open up the datasheet. I searched by the part number to find it.
-
Today I took "special tools" with me from work.
These two (slightly rusty) wrenches are only used to screw the 4mm safety bushes properly.
You insert the thinner one at the front and turn it slightly until it clicks into place, then use the larger one to tighten the nut.
You can also do this with needle-nose pliers, but it's much easier and tighter this way.
-
Update:
No update, I´m a lazy boy... ;)
Or better, time is not enough, but it is not forgotten.
Several projects are currently in the pipeline:
- Voltage Reference, installation in the housing and ageing before it is sent in for calibration.
- 1kHz ultra low distortion oscillator:
Installation in the housing, testing.
- Arduino Curve Tracer:
Finish assembling, then test.
- ENOB test environment:
Filter boards designed, still need to be assembled and tested.
- New HMI for Hioki 5Mhz LCR:
So far only collected ideas
And a lot more....
As I said, lazy boy. ;)
-
On the bright side, you'll probably have your SDS3000XHD by the time you finish all that. 🤣🤣
-
Do you know the foot massage scene in pulp fiction?
The final punchline where Vincent Vega had Julius by the balls and how Julius reacted to it ? ;)
I´m Julius.... 8)
-
Nah, I don't remember. Maybe I'll watch the movie again and get back to you. I could probably watch it a couple times before.......................... 😉