On Wed, 17 May 2017 09:14:56 +1000, Clifford Heath <clifford.heath@gmail.com> wrote:>On 16/05/17 06:17, John Larkin wrote: >> On Mon, 15 May 2017 19:51:11 +0100, "Kevin Aylward" >> <kevinRemovAT@kevinaylward.co.uk> wrote: >> >>>> "bitrex" wrote in message news:xW2SA.99882$jo6.643@fx33.iad... >>> >>> it... >>>> >>>>> Is your object to drive a length of coax? If it's terminated, you can >>>>> just ignore the cable entirely. Or poke in a Spice transmission line >>>>> to explore mismatch consequences. >>>> >>> >>>> Ya, exploring the "mismatch" consequences I'd like to do. How do I do that? >>>> I haven't used Spice transmission line models before... >>> >>> Well... in SS, you just place a transmission line symbol from the lib, and >>> double-click on it to pop up its setup box. :-) >> >> LT has a transmission line. It's fully symmetric, like an unshielded >> twisted pair. And it acts as if it has an ideal transformer inside; it >> tolerates any common-mode difference between ends. In fact, it makes a >> handy ideal transformer. >> >> To simulate an actual twisted pair, I use three of the LT Spice TXLINE >> things. >> >> https://dl.dropboxusercontent.com/u/53724080/Circuits/TxLines/Twist_1.jpg >> >> Probably the middle one should have a different prop delay. > >Is that to handle the shield? How would you model UTP?That model is about right for shielded CAT6. UTP is sort of shielded by the universe, in the sense that it still has odd and even modes. You don't care much about the even mode in Ethernet, since the signals are transformer coupled and nicely balanced on both ends. So a single LT Spice TLINE or LTLINE would be an OK model for UTP. LTLINE still wouldn't get the skin loss right. I have an application that uses a shielded twisted pair but has unbalanced terminations, so both the odd and even propagation modes matter some. I'm trying to get the customer to send me a hank of the actual cable, so I can TDR it and tweak my model. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Fast buffer idea
Started by ●May 14, 2017
Reply by ●May 16, 20172017-05-16
Reply by ●May 16, 20172017-05-16
>I was doing femtosecond stuff in 1970.Do tell. "Femtosecond stuff" such as reading the newspaper by 600 THz light, maybe? I'm pretty sure I'm the only one in this group who has actually built electronics that respond on fs time scales (optical antennas coupled to metal-insulator-metal tunnel diodes--basically crystal radios for light). Picosecond lasers were the state of the art in the late '80s. One or two groups had used pulse compression to make pulses at the 10-20 fs level as a hero experiment by then, but that sure wasn't happening in 1970. So what "femtosecond stuff" were you doing in 1970? Cheers Phil Hobbs
Reply by ●May 16, 20172017-05-16
John Larkin <jjlarkin@highland_snip_technology.com> wrote:> On Tue, 16 May 2017 21:21:49 GMT, Steve Wilson <no@spam.com> wrote: > >>John Larkin <jjlarkin@highland_snip_technology.com> wrote: >> >>> On Tue, 16 May 2017 20:33:54 GMT, Steve Wilson <no@spam.com> wrote: >>> >>>>John Larkin <jjlarkin@highland_snip_technology.com> wrote: >>>> >>>>> On Tue, 16 May 2017 18:55:54 GMT, Steve Wilson <no@spam.com> >>>>> wrote: >>>> >>>>>>But you have already forgotten. Remember the tiny pcb you made >>>>>>where you didn't have enough room to add parasitic suppression >>>>>>resistors? >>>>> >>>>> When you're doing fast stuff, and every picosecond is worth money, >>>>> adding any resistor in the signal path, or wasting any space, is a >>>>> defeat. >>>> >>>>Nice marketing, poor engineering. >> >>> Design some picosecond stuff yourself. And show us. And don't be an >>> asshole. >> >>I was doing femtosecond stuff in 1970. Your insult about being an >>asshole has no effect. Stop being a jerk. >> >>>>When you are trying to ship fast stuff, and it oscillates, that is a >>>>defeat. >>> >>> It got fixed and we got paid. >> >>So what heppened to your picoseconds? >> >>>>You have to fix it, and you had no room. >>>> >>>>You should have expected it and solved it before making lots of >>>>pcbs. Coupled high speed transistors or facing long leads is a >>>>potential disaster. >>>> >>>>I'll bet you check now. >> >>> I've been doing picosecond electronics for decades, and sold a ton >>> of it... more as time goes on. Sometimes things happen on the first >>> PCB rev of a difficult design. We fix it. No big deal. >> >>As I said, you should have expected it, and you did not. So much for >>your claims of picosecond electronics. >> >>In fact, nothing you do is in picoseconds. Perhaps tenths of a >>nanosecond, but not much more. False claims, but they can make a lot >>of money. > > > https://dl.dropboxusercontent.com/u/53724080/Optos/95_ps_optical_pulse. > jpg > > https://dl.dropboxusercontent.com/u/53724080/Circuits/Fast_Stuff/100_ps > _steps.jpg > > https://dl.dropboxusercontent.com/u/53724080/Circuits/Fast_Stuff/NB7_St > eps.jpg > > https://dl.dropboxusercontent.com/u/53724080/Circuits/Fast_Stuff/DSRD_n > eg-2KV.JPGLike I said, tenths of a nansecond stuff, with a lot of jitter. Good marketing, poor engineering. But very few of your customers can tell the difference.
Reply by ●May 16, 20172017-05-16
pcdhobbs@gmail.com wrote:>>I was doing femtosecond stuff in 1970. > > Do tell. "Femtosecond stuff" such as reading the newspaper by 600 THz > light, maybe? > > I'm pretty sure I'm the only one in this group who has actually built > electronics that respond on fs time scales (optical antennas coupled > to metal-insulator-metal tunnel diodes--basically crystal radios for > light). > > Picosecond lasers were the state of the art in the late '80s. One or > two groups had used pulse compression to make pulses at the 10-20 fs > level as a hero experiment by then, but that sure wasn't happening in > 1970. > > So what "femtosecond stuff" were you doing in 1970? > > Cheers > > Phil HobbsI have a great deal of respect for your engineering capability. It is probably second to none in optics. I was in a different industry. I designed products for the hard disk industry. I made timing measurements in production test to 100fs. This was essential to meet the accuracy and linearity requirements needed. I was the only person in the industry who could do this. I made millions. There is a big difference between 100fs risetime and several tens of picoseconds of rms jitter. You should make some tutorials to make this clear to people. You have posted a number of simple equations that describe this. I chide Larkin on his "picosecond electronics". Do you accept his claims? I thought not.
Reply by ●May 16, 20172017-05-16
On 14 May 2017 17:58:54 -0700, Winfield Hill <hill@rowland.harvard.edu> wrote:>bitrex wrote... >> >> On 05/14/2017 03:28 PM, Kevin Aylward wrote: >> >>> I haven't examined the effectiveness of the additional boost circuit, >>> but it is important to make sure the basic design is optimised first. >>> Typically, something like a 1:4 build up in current from the first bip >>> follower to the second bip follower. In an asic design I might use the >>> same current and use multiple devices in parallel. I would suggest >>> trying an active current source to the input followers, to get the drive >>> to the second follower, and remove the boost circuit. Use standard >>> configurations, until you can prove that a bright idea is truly better. >> >> The problem with a simple active current source to the input followers >> is quiescent current draw - if you optimize the follower current source >> for the gain you need at the highest frequency the buffer draws a huge >> helping of quiescent current sitting around doing nothing. > > Devices that are off, or mostly off, are slow to turn on. > You suffer from a t = C V / i delay time. SPICE fails > to give you the proper value for C, and the delay for i. > This is especially true if you invoke any MOSFET models. > > You can make SPICE devices do all kinds of amazing things. > This is NOT to say the real device can do the same thing.SPICE doesn't "fail", the crappy model failed. "This is especially true if you invoke any [discrete] MOSFET models." Chip foundry models are superb, and have never failed me. ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | Thinking outside the box... producing elegant solutions. "It is not in doing what you like, but in liking what you do that is the secret of happiness." -James Barrie
Reply by ●May 16, 20172017-05-16
On Tue, 16 May 2017 17:27:24 -0700, Jim Thompson <To-Email-Use-The-Envelope-Icon@On-My-Web-Site.com> wrote:>On 14 May 2017 17:58:54 -0700, Winfield Hill ><hill@rowland.harvard.edu> wrote: > >>bitrex wrote... >>> >>> On 05/14/2017 03:28 PM, Kevin Aylward wrote: >>> >>>> I haven't examined the effectiveness of the additional boost circuit, >>>> but it is important to make sure the basic design is optimised first. >>>> Typically, something like a 1:4 build up in current from the first bip >>>> follower to the second bip follower. In an asic design I might use the >>>> same current and use multiple devices in parallel. I would suggest >>>> trying an active current source to the input followers, to get the drive >>>> to the second follower, and remove the boost circuit. Use standard >>>> configurations, until you can prove that a bright idea is truly better. >>> >>> The problem with a simple active current source to the input followers >>> is quiescent current draw - if you optimize the follower current source >>> for the gain you need at the highest frequency the buffer draws a huge >>> helping of quiescent current sitting around doing nothing. >> >> Devices that are off, or mostly off, are slow to turn on. >> You suffer from a t = C V / i delay time. SPICE fails >> to give you the proper value for C, and the delay for i. >> This is especially true if you invoke any MOSFET models. >> >> You can make SPICE devices do all kinds of amazing things. >> This is NOT to say the real device can do the same thing. > >SPICE doesn't "fail", the crappy model failed.Emendment: LTspice may lie to you and show an unrealistic speed... unless you understand Spice and set the control parameters properly, AND select Solver=Alternate.> >"This is especially true if you invoke any [discrete] MOSFET models." > >Chip foundry models are superb, and have never failed me. > > ...Jim Thompson...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | Thinking outside the box... producing elegant solutions. "It is not in doing what you like, but in liking what you do that is the secret of happiness." -James Barrie
Reply by ●May 16, 20172017-05-16
On Tue, 16 May 2017 23:58:34 GMT, Steve Wilson <no@spam.com> wrote:>John Larkin <jjlarkin@highland_snip_technology.com> wrote: > >> On Tue, 16 May 2017 21:21:49 GMT, Steve Wilson <no@spam.com> wrote: >> >>>John Larkin <jjlarkin@highland_snip_technology.com> wrote: >>> >>>> On Tue, 16 May 2017 20:33:54 GMT, Steve Wilson <no@spam.com> wrote: >>>> >>>>>John Larkin <jjlarkin@highland_snip_technology.com> wrote: >>>>> >>>>>> On Tue, 16 May 2017 18:55:54 GMT, Steve Wilson <no@spam.com> >>>>>> wrote: >>>>> >>>>>>>But you have already forgotten. Remember the tiny pcb you made >>>>>>>where you didn't have enough room to add parasitic suppression >>>>>>>resistors? >>>>>> >>>>>> When you're doing fast stuff, and every picosecond is worth money, >>>>>> adding any resistor in the signal path, or wasting any space, is a >>>>>> defeat. >>>>> >>>>>Nice marketing, poor engineering. >>> >>>> Design some picosecond stuff yourself. And show us. And don't be an >>>> asshole. >>> >>>I was doing femtosecond stuff in 1970. Your insult about being an >>>asshole has no effect. Stop being a jerk. >>> >>>>>When you are trying to ship fast stuff, and it oscillates, that is a >>>>>defeat. >>>> >>>> It got fixed and we got paid. >>> >>>So what heppened to your picoseconds? >>> >>>>>You have to fix it, and you had no room. >>>>> >>>>>You should have expected it and solved it before making lots of >>>>>pcbs. Coupled high speed transistors or facing long leads is a >>>>>potential disaster. >>>>> >>>>>I'll bet you check now. >>> >>>> I've been doing picosecond electronics for decades, and sold a ton >>>> of it... more as time goes on. Sometimes things happen on the first >>>> PCB rev of a difficult design. We fix it. No big deal. >>> >>>As I said, you should have expected it, and you did not. So much for >>>your claims of picosecond electronics. >>> >>>In fact, nothing you do is in picoseconds. Perhaps tenths of a >>>nanosecond, but not much more. False claims, but they can make a lot >>>of money. >> >> >> https://dl.dropboxusercontent.com/u/53724080/Optos/95_ps_optical_pulse. >> jpg >> >> https://dl.dropboxusercontent.com/u/53724080/Circuits/Fast_Stuff/100_ps >> _steps.jpg >> >> https://dl.dropboxusercontent.com/u/53724080/Circuits/Fast_Stuff/NB7_St >> eps.jpg >> >> https://dl.dropboxusercontent.com/u/53724080/Circuits/Fast_Stuff/DSRD_n >> eg-2KV.JPG > >Like I said, tenths of a nansecond stuff, with a lot of jitter. > >Good marketing, poor engineering. But very few of your customers can tell >the difference.The NIF timing system had about 2000 endpoints spread around the biggest laser in the world. Jitter across the system was about 3 ps RMS, or maybe it would have been better if we'd had better oscilloscopes. https://dl.dropboxusercontent.com/u/53724080/NIF3.jpg We could do better today. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by ●May 16, 20172017-05-16
>I chide Larkin on his "picosecond electronics". Do you accept his claims?>I thought not.You really shouldn't let your temper lead you into putting words in people's mouths. I have a few of JL's products in my lab. My favourite is the P400 digital delay generator, which I use a fair amount--it's my go-to pulse generator for everything. It's similar to an SRS delay generator I used to have except that its jitter is a good 30 dB less (~5 ps vs ~200 ps). That sort of improvement doesn't happen by accident. You claim hundreds-of-femtoseconds jitter numbers for hard disks in 1970. Even if that were so, that's "sub-picosecond" and not "femtosecond". However the state of the art in hard drives in 1970 was nowhere near that jitter level. The 1973 IBM "Winchester" drive had a data transfer rate of less than 1 Mb/s. Sub-picosecond jitter? Really? Like 120-odd dB down? I seriously doubt that available test equipment could have made that measurement, even if by some miracle it was that good. How did you measure it? Cheers Phil Hobbs
Reply by ●May 16, 20172017-05-16
On 05/16/2017 02:47 PM, Jan Panteltje wrote:> On a sunny day (Tue, 16 May 2017 08:12:13 GMT) it happened Jan Panteltje > <pNa0nStpealmtje@yahoo.com> wrote in <ofec8t$1a9r$1@gioia.aioe.org>: > >> <bitrex@de.lete.earthlink.net> wrote in <zsnSA.28967$9o7.3799@fx12.iad>: > >>> Yes, the basic topology is an _excellent_ performer at low quiescent >>> bias current, low frequency, and into no load! ;-) >> >> Na, be a good sport, if I have time tonight I will add a 75 Ohm load, >> and rebuild it so the driver PNP is making thermal contact with the output NPN (for each half). >> I hope that will fix any thermal tracking. >> The above setup I build faster than you spicers could enter it. >> AND it is real, unlike in the matrix.. >> ;-) > > So.... > been doing some testing, 75 Ohm load: > http://panteltje.com/pub/some_detail_IMG_6194.JPG > > 2 signal sources, laptop 20 kHz and 60 MHz square wave oscillator module (on small PCB borrowed from other HF project). > output loaded with 75 Ohm coupled with a capacitor as I am testing with single rail lab supply between 5 and 10V. > and I decoupled the supply line too this time. > > 20 kHz in 75 Ohm: > http://panteltje.com/pub/input_output_20kHz_75_Ohm_load_IMG_6193.JPG > input and output guess who is who (I know). > > Since I do not have one of those defective scopes, the ones that only work every now and then, > sometimes call 'sampling scopes [1]' by their users, I use a trick > http://panteltje.com/pub/60MHz_square_wave_in_antenna_out_plus_75Ohm_load_IMG_6192.JPG > send the output to a piece of wire, and via LAN start raspberry that has antenna in attic sdr_rtl and my spectrum analyzer (I wrote it) > > to see what comes out as harmonics.... of the 60 MHz square wave I feed in: > http://panteltje.com/pub/testing_second_harmonic_of_60MHz_squarewave_IMG_6191.JPG > base at 60 MHz, second at 120 MHz, third at 180 MHz and a weak fourth at 240 MHz no problem finding with sdr gain at minimum. > > Hope this helps. > > Any more questions please build it yerself, it is not that hard. > > Thermal runaway? No sign of it, not with these voltages and 75 Ohm load. > So .. else glue some trannies together so they track. > nuf said. > > [1] I do have one but it is slow:m > http://panteltje.com/panteltje/pic/scope_pic/ > > >Ideally one could use one of those monolithic multi-transistors on a single die, like the CA3096...it doesn't seem like you can get 'em for any reasonable price anymore, though. They have them direct from China on eBay. I'm a little leery of that option...burned before! I also had an idea for a faster variation on the LM13700 OTA using a couple transistor array ICs and a Burr Brown current source/mirror chip. It would "only" cost 15 bucks...:(
Reply by ●May 16, 20172017-05-16
On 05/16/2017 04:16 AM, Jan Panteltje wrote:> On a sunny day (Mon, 15 May 2017 15:33:15 -0400) it happened bitrex > <bitrex@de.lete.earthlink.net> wrote in <0onSA.73595$jO6.6010@fx41.iad>: > >> On 05/15/2017 03:13 PM, Jan Panteltje wrote: >> >>> Well that is same as from black to white, works fine tough. >>> that is just the PNP conducting pulling the output. >>> This circuit has been used for video by me many many times. >>> Never ever failed. >> >> Ya I'm sure it works fine for composite video. If the risetime off the >> porches is a little sluggish I don't think it matters much; there isn't >> anything timing-critical to that transition as its just setting up the >> luminance "data." Might make the edges a little fuzzy? Like "vintage warmth" > > It (the amplifier) is > 5MHz bandwidth, > this is analog NTSC send down from the drone at 5.something GHz. > You do NOT want faster risetimes etc in analog PAL / NTSC video. > It only adds crap (plenty HF noise in that FM link) . > You are no video man it seems, you talk about 'porches' nobody in video does. >No, I've never really designed or worked with analog video amps/RF modulators/NTSC etc. I'm in my 30s, for most of my adult life "video" means "you plug in the magic (digital) cable and it works." I only know the terms and a little bit about that technology, like 8 track players and vacuum-tube computers, from what you might call "books of historical interest" ;-)
