Recent Posts

31

Projects, Designs, and Technical Stuff / Re: TVS Diode Usage and Diode breakdown voltage

« Last post by injineerka on Today at 08:48:35 am »

Thanks for the reply. It is current limited upstream. Yes it is indeed schottky diode, and the main reason was the small size and low Vf. But I might need to reconsider it again and look for smaller package of standard diode.
Thanks again, it was helpful insight.

32

Projects, Designs, and Technical Stuff / Re: T48 EPROM Programmer : Recommended adapters ???

« Last post by dazz1 on Today at 08:44:57 am »

The T48 kits with the full range of 40+ adapters are likely to include a lot of adapters I will never use. 
I suspect there is a middle point of a kit that includes the most commonly used adapters.

33

General Technical Chat / Re: It's "sodder", not "soLder", damnit! (for us USAians)

« Last post by Gyro on Today at 08:39:39 am »

Oh, FFS, not another one. This has been done to death in many threads over the years. I suggest you visit the outside world rather than harping on about your US centric view (you clearly haven't, judging from your posts on power outlets, switch orientations etc.) . Maybe you can message Trump and get him to decree the 'great standard American Sodder' spelling (rather apt), it's great!, it's the greatest! the greatest in history!

Solder is pronounced as it is is spelled:  S o l d e r. Just as with  A l u m i n i u m  and  N u c l e a r.  Just because some US solder wick [Edit: actually 'desoldering braid] manufacturer has used slang in its product name proves nothing.

Now, can we please stop cluttering the forum with these utterly pointless bored OP threads and discuss technical things!

34

Test Equipment / Re: Uni-T UTD1050DL oscilloscope firmware dump needed

« Last post by Step12 on Today at 08:35:30 am »

Good afternoon!
I read your post and wanted to ask for your help.
Some guys here have already done me a huge favor, but I still have a few questions...
Could you please tell me if you connected directly to the 9J2 connector using an FTDI adapter?
If it’s not too much trouble, could you enlighten me on how to do that?
Thanks in advance, Stepan!

35

Beginners / Re: Is my understanding & mental map of Types of Circuit Analysis correct?

« Last post by isitayush on Today at 08:32:14 am »

The intuition you need to learn is, for any given circuit, which analysis are necessary. Remember, in the real world with physical lab instrumentation (NOT spice simulations), setting up equipment and performing testing takes time and money, wasted if the analysis is unnecessary.  Even unnecessary spice simulations can waste time.

Thank you, I understand what you mean. My goal with this "axes" was to build an analysis taxonomy. I know that If I'm testing a band pass filter in the lab, I would go for (4) Small Signal, Frequency Domain, AC / Perturb analysis (.AC) instead of say (6) Large Signal, Time Domain, AC / Perturb (.TRAN) unless I think I need it.

Your categorical 3-axes matrix fails to apply the analyses to the problem.  In effect, you are performing a meta-analyses, something which engineers don't usually do, but theoreticians do.

I wrongfully said "There are 8 models", I apologise. It should be "There aren't 8 models" and I've corrected my mistake. My 3-axes aren't fundamental dimensions of reality living in cartesian space, they're more of a meta-dimensions that help describe how many total mathematical analyses someone (simulator or human) can ever hope to perform for any given circuit. It does not tell you which quadrant/analysis, you must choose, It's up to you. It only tells you, how many total exist regardless of the circuit. To choose a quadrant/analysis, you must answer these questions.

Code: [Select]

1) Axis of Non-Linear Device Modelling / Axis of Linearity: Do we keep the full non-linear approximations for non-linear devices (i.e. a large signal analysis) or Do we use the first order talyor approximation for non-linear devices (small-signal analysis)
2) Axis of Mathematical Representation:  Do we represent the circuit in time domain (transient analysis) or in frequency domain (frequency analysis)?
3) Axis of Input Excitation: Is my input an AC signal or a DC signal?


36

Projects, Designs, and Technical Stuff / Re: TVS Diode Usage and Diode breakdown voltage

« Last post by injineerka on Today at 08:18:22 am »

I know that it is unidirectional. What i meant was that it should be bidirectional TVS instead of unidirectional.

37

Test Equipment / Re: Keithley DMM7510 Hidden & Undocumented Features

« Last post by Richard W. on Today at 08:16:10 am »

Thanks  I wonder, if some of them work on my 2601A as well...

display.OFF/display.ON would be nice if they work (for lengthy runs, where nobody watches the display. A simple 34401A has this feature...)

also the "fan" functions could be interesting, "gettemp" to see if its already stabilized and the ability to override the speedsettings.
In some cases its switching between two levels in a 5sec interval. Thats quite annoying if you sit right next to it.
Also for times when less accuracy but more power is necessary, one could set the fan to 100% continuously to keep the device cooler.

I already found out some of the math functions, like "math.mod()" for modulo operations and math.ceil(), math.floor() for rounding up and down work.

38

Test Equipment / Re: Possible reasons *not* to get an analog oscilloscope

« Last post by Aldo22 on Today at 08:00:43 am »

@tautech
@CJSouthern
@2N3055

Of course, the scope is superior when it comes to measuring and visualizing AC. That’s absolutely clear.
But often, for the sake of convenience, you’ll start with a handheld DMM, and then a DMM with AC&DC display might alert you that you should take a closer look at the AC component with a scope.
It’s not even a question of whether the DMM’s AC RMS measurement is accurate in this case—any value >0 indicates that you should take a closer look.
It’s not perfect, just a small advantage.
For me, this dual display has been helpful on several occasions.

By the way, I write AC&DC instead of AC+DC because I’m referring to the dual display.
AC+DC is the value that corresponds to the RMS measurement of an oscilloscope in DC coupling.

39

News/Suggestions/Help / Re: Getting a lot of "502 Bad Gateway" errors

« Last post by gnif on Today at 07:55:45 am »

CF failed to identify a massive attack against our infrastructure and left us with no tools to filter it. This was a few years ago, here is hoping that the situation has improved.

CF is supposed to automatically block huge DDOS attacks (gigabit volume). That was the original reason for CF. But - as we found too - it does not block "small" attacks which are enough to trash your site (and ours). For those you have to use "proof of work" like Precursor or, if your server bw is enough, set up Anubis on the server.

Proof of work serving is low server load but very high client load, which is why it works.

I spent months blocking IP ranges and now have almost 1000 blocked, like e.g. all of Hetzner, Digital Ocean, etc. That is a good idea anyway (some will moan because they are on VPNs terminating on these farms). I also block Russia and China (which I know you don't want to). Historically we got a lot of attacks from some guys in Germany... hence the German server farms. But this last attack, a huge rented residential botnet, 50% US and 50% 3rd World, no chance of blocking that.

Yes, we have a solution for that already but it was not designed to work with CF. I can adjust it if we need to and put it back in place. Again though the specifics of the implementation are not going to be shared for operational security reasons.

Our block lists for DC owned subnets span hundreds of thousands of IPs, and are automatically generated based on external sources of information daily (Attached non-exhaustive sample). We do not block them, but rather force a proof of work.

40

Projects, Designs, and Technical Stuff / Re: PCBGolf: $1,000 PCB layout challenge

« Last post by hans on Today at 07:52:08 am »

Without mechanical constraints on connectors, this exercise is a bit pointless imo. Otherwise, indeed, I would put the barreljack on middle of the PCB, argue a bunch of 0402s get assembled first and nobody asked the barreljack needs to be mounted flush on the PCB. There are so many rules to bend.. the PCB artwork is often not the problem.