Voltage Spike

Please post a schematic

Can be anywhere, but needs to have fast shipping time to US. Qtys are low 10 to 100 for proto with max volume in the 1000s Id like the lead time to be 2 to 3 weeks

Anybody know of other PCB suppliers other than Sierra circuits that offer turnkey Assembly that's far for a reasonable price?

Mainly because I got a pre-engineering associates and then got a degree in EE

In high school I could get college credit for many of the classes, such as trig and calculus. I could also take courses from universities over CC tv and get both HS and college credit. I had enough credits for 1/2 of a BS and it still took me 5 years to graduate.

And we had no recourse because it's TI

I've had some of the auto zero/chopping opamps give me problems. We had to do a product recall because of it.

@tobalt TI is not good, I've had issues with TI parts. IMO the analog market has been 'researched out' There aren't a lot of areas to expand with new parts and the areas that have been done the gains are harder to get.

@NickAlexeev I don't do the bluetooth, but we do use silabs and it works well. The tools provided are nice. We did try a STM32 bluetooth chip STM32WB55RGV7 but that chip is really picky with the noise on 3.3V VCC line and resets sometimes because they didn't design the silicon right. I think they have a fix for it but it was surprising coming from ST. The STM32WB's work better but are more pricey

@MattD EE is like picking a major again.

If you only have one thread then use bare metal

An RTOS essentially lets you time processes and gives you real time control over threads.

And you have a processor that can run it.

@misk94555 An RTOS will give you more complexity but won't give you better timing than bare metal. If you need to leverage library and still be fast or multicore then use RTOS IMO. If you can program it with bare metal then use that. Use linux if you don't need timing that is super fast like 10's of msec.

@NickAlexeev I'll need it for a few projects so I'd like to buy one. I guess the calibration of the PCB makes them so expensive?

Anybody know of an ethernet compliance tester (10/100) that I can get for less than 2k? Something like this: testequipmentdepot.com/…

I'm wondering if there are any drawbacks.

The paper on these sodium batteries is impressive. They outpreform lithium ion. eurekalert.org/news-releases/1050283

I like bosch stuff, I don't think it does anything like that...

Just the script

I've done a few scripts, not with zip sad thing is AFAIK you can't write another output container

Any conceivable way to automatically zip files in altium 24 with export?

Yeah, at 100kHz, I wouldn't think that making an LCR meter with most micros with an integrated ADC/DAC. The STM32L432 does 5MSPS (but that might be combined sample rate, not sure) max so that leaves a lot of room for a 100kHz signal. I've done some things with detecting inductance and basically the only thing you need is amplitude. If you have many frequencies it's much easier to estimate.

@misk94555 Really you need to know what will be on top of DC, there will be variation of fluid flow and it would be best if you characterize that before you build it, perhaps experimentation will give you the best measurements. The last thing to worry about is simultaneous sampling, and how much the signal changes with time. If the signal changes too much with time it can add 'noise' to the subtraction.

This is dependent on local conditions so you would need to determine what rage of pressure values the sensor will see and see if differential or pseudo diff will work better

@misk94555 That's more of a mechanical engineering question as it deals with fluids. I don't know about your specific problem but usually differential measurements are employed when you either have lots of noise impacting the sensor or not doing the subtraction at the sensor will cause the instrument to saturate and the measurement is lost.

@NiallC Do they have the migration que setup at DIY?

@NiallC. Are you a mod there?

@NiallC. Sure thing! Just because it's an industrial furnace doesn't mean its not a hobbyist application. Looks more like a guy in his backyard

@W5VO Don't forget you are paying for calibration

Sorry that message was garbled, I'll fix it when I get back to a computer

And LCR meters, they use a frequency to determine part valuesYou should look at smart tweezers

What is your speed, bandwidth\waveform requirments?

unfortunate it will take some reading, there is a lot of functionality so it may be difficult to get working.

This is one I've used in the past. analog.com/en/products/ad9833.html

But not always

@JohnRennie It's a little unclear what you are asking. DDS is usually used in digital logic to generate waveforms. No oscillators contain micros, micros can contain oscillators.

@JohnRennie The CMOS oscillators are some kind of resonator on silicon or a crystal with an amplifier attached that gives square wave output. Other crystals need an external excitation circuit which is usually on the microprocessor. You can look at both types here: digikey.com/en/products/filter/oscillators/…

@NickAlexeev Thanks for the info, that is good to know

@Lundin I have talks with Avnet and future on a regular basis. Both have a good office in the area and have good reps\support engineers. I'm surprised that you have had that experience.

I did get an RMA and free shipping but I'm not happy about the invoice.

So I order a few feet of EMI gasket from digikey and they send me a spool with +1000ft. This stuff is expensive, a few dollars a foot. So the spool is worth a few thousand dollars, I figure that it would be good to get it back to them. So I tell them and they serve me an invoice like it's my fault! I told them that I really didn't like that so I'll see what they say.

@NickAlexeev I think I'm going to roll my own and do a test cable from PCBway with their FFC services. The price is about 1$ a unit which is great for testing and if we did go to final product with the design, I'm sure we could get it for as cheap as the unshielded molex we have. The LVDS cable was about 4$ at 1k qty's so that is pricey.

@Lundin You can get custom FFC cables done with different EMC shielding materials over the coverlay, the NRE's can be a bit pricey (maybe 250-500 more for qty 50) . The materials must be tied in with copper/vias. What I have is an FFC with neither ground plane or shielding.

@NickAlexeev Thanks for that, that is good to know. Yeah, I had some ferrites on the way to test, this wouldn't be a great thing to add to the BOM as it's fairly custom. I'd like to short out the HF at the PCB with three terminal low inductance caps from johnson dielectric. I suppose also adding some impedance on the lines would be appropriate.

I think you might also be able to find a custom o-ring supplier that could make what you want, I'm not sure which one would be best and specializes in micro gaskets but I know most gaskets can be custom made with very low energies or free

Frequency emissions?

The schematic is from flastad, I can't tell which voltages you are applying to the circuit or the model you are using. Use spice instead

@RuviandeCésaro What simulator are you simulating this on. Can you give a schematic and the values that you are trying to simulate? Why aren't you using a spice package.

@ShikShoxF1 Look at other h-bridge schematics to see how they draw them. A good schematic is readable. Text is not drawn over wires or components. VCC's go up over components, GNDS go down. NPN's and PNP's are usually drawn with the emitter down. If there are two many wires, using net names and separating wires is advisable.