From: krw on
On Sat, 12 Sep 2009 12:25:31 -0700, Jim Thompson
<To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> wrote:

>On Sat, 12 Sep 2009 11:55:25 -0700, Joerg <invalid(a)invalid.invalid>
>wrote:
>
>>Jim Thompson wrote:
>>> On Sat, 12 Sep 2009 11:47:08 -0700, Joerg <invalid(a)invalid.invalid>
>>> wrote:
>>>
>>>> Jim Thompson wrote:
>>>>> On Sat, 12 Sep 2009 11:25:35 -0700, Joerg <invalid(a)invalid.invalid>
>>>>> wrote:
>>>>>
>>>>>> Jim Thompson wrote:
>>>>>>> On Sat, 12 Sep 2009 10:01:22 -0700, Jim Thompson
>>>>>>> <To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> wrote:
>>>>>>>
>>>>>>>> On Sat, 12 Sep 2009 08:51:21 -0700, Joerg <invalid(a)invalid.invalid>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> John Larkin wrote:
>>>>>>>>>> On Fri, 11 Sep 2009 16:38:42 -0700, Joerg <invalid(a)invalid.invalid>
>>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>>> John Larkin wrote:
>>>>>>>>>>>> On Fri, 11 Sep 2009 15:33:08 -0700, Joerg <invalid(a)invalid.invalid>
>>>>>>>>>>>> wrote:
>>>>>>>>> [...]
>>>>>>>>>
>>>>>>>>>>>>> One issue I found with RF devices was that you can sometimes only obtain
>>>>>>>>>>>>> small signal simulator models that aren't very helpful with pulse apps,
>>>>>>>>>>>>> plus require software that costs more than your Volkswagen. Even with
>>>>>>>>>>>>> big LDMOS parts I had that happen. One company (PolyFet) stuck out and
>>>>>>>>>>>>> provided SPICE models. So the big competitor will be left out ... sorry,
>>>>>>>>>>>>> NXP.
>>>>>>>>>>>> I don't simulate this stuff, I test. Most microwave parts are
>>>>>>>>>>>> characterized small-signal, S-params. Spice models are very rare in
>>>>>>>>>>>> this business. I know things about some of these parts that the
>>>>>>>>>>>> manufacturers don't know.
>>>>>>>>>>>>
>>>>>>>>>>> Lately I have simulated more, and then straight to layout. The one
>>>>>>>>>>> that's in fab right now is a white-knuckle ride because I am using a
>>>>>>>>>>> device in a weird way. Usually pans out though.
>>>>>>>>>>>
>>>>>>>>>>> What often amazes me is how close to abs max people (and sometimes app
>>>>>>>>>>> notes) bias RF devices. When it says 7V they bias them to 5V or 6V on
>>>>>>>>>>> the collector/drain. Some day a huge pulse finds its way into the input,
>>>>>>>>>>> almost saturates the device, then it lets go, the inductor shoots up ...
>>>>>>>>>>> phut.
>>>>>>>>>> I've run 7-volt-rated MESfets at 18 volts, and 2 volt RF schottkies at
>>>>>>>>>> 6. I sometimes test parts to destruction and then back off some. The
>>>>>>>>>> performance is often worth the small risk.
>>>>>>>>>>
>>>>>>>>> I've seen that. But if you are a consultant and this goes into some
>>>>>>>>> aircraft you can't design like that. Even with an indemnification clause
>>>>>>>>> that can cause lots of grief should it go wrong in only one case. It's
>>>>>>>>> got to be by the book.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> Lots of RF devices seem to have voltage ratings that assume an RF tank
>>>>>>>>>> in the output, so that the actual instantaneous drain excursion is 0
>>>>>>>>>> to twice Vcc. The RF guys are really, really terrible when it comes to
>>>>>>>>>> realtime specs. You're lucky to get any DC curves. More often there's
>>>>>>>>>> an app circuit with an input match, an output match, and a gate bias
>>>>>>>>>> trimpot.
>>>>>>>>>>
>>>>>>>>> All they really offer in many cases is a set of S-parameters. Since I
>>>>>>>>> design a lot of pulse circuitry I almost do a rain dance every time
>>>>>>>>> there is just a snippet of SPICE data.
>>>>>>>> Somewhere in my bag-o-tricks is an S-Parameter-to-Spice-Model
>>>>>>>> converter (and vice-versa).
>>>>>>>>
>>>>>>>> When I find it I'll post.
>>>>>>>>
>>>>>>>> ...Jim Thompson
>>>>>>> All my S-Parameter to Spice stuff has been updated on the SED page of
>>>>>>> my website... except a good article by Microcap... get that from their
>>>>>>> website, so I don't step on any copyright toes ;-)
>>>>>>>
>>>>>> Thanks. I assume you mean this link:
>>>>>>
>>>>>> http://www.analog-innovations.com/SED/CreateS-ParameterSUBCKTinPSpice.pdf
>>>>>>
>>>>>> Now I have to learn QuickBasic, whatever that is ;-)
>>>>> There are four articles there, each beginning (in the listing),
>>>>> "S-Param...", all of which are elucidating.
>>>>>
>>>> Unless I understand something wrong the first three are for extracting
>>>> S-parameters from SPICE or PROBE (Orcad's version of SPICE).
>>>>
>>>> But maybe I'll read them again and some more elucidation will come :-)
>>>
>>> Most math is bilateral ;-)
>>>
>>
>>Ok, true, I was lazy here. Thought about some SW-routine where you slap
>>the datasheet on a scanner, press magic button, out comes SPICE
>>sub-circuit :-)))
>
>I can do that for you... for a fee ;-)

You want Joerg pressing your magic buttons? ...for a fee? Ew!
From: John Larkin on
On Sat, 12 Sep 2009 13:14:20 -0700, Joerg <invalid(a)invalid.invalid>
wrote:

>John Larkin wrote:
>> On Sat, 12 Sep 2009 11:15:41 -0700, Joerg <invalid(a)invalid.invalid>
>> wrote:
>>
>>> Jim Thompson wrote:
>>>> On Sat, 12 Sep 2009 08:51:21 -0700, Joerg <invalid(a)invalid.invalid>
>>>> wrote:
>>>>
>>>>> John Larkin wrote:
>>>>>> On Fri, 11 Sep 2009 16:38:42 -0700, Joerg <invalid(a)invalid.invalid>
>>>>>> wrote:
>>>>>>
>>>>>>> John Larkin wrote:
>>>>>>>> On Fri, 11 Sep 2009 15:33:08 -0700, Joerg <invalid(a)invalid.invalid>
>>>>>>>> wrote:
>>>>> [...]
>>>>>
>>>>>>>>> One issue I found with RF devices was that you can sometimes only obtain
>>>>>>>>> small signal simulator models that aren't very helpful with pulse apps,
>>>>>>>>> plus require software that costs more than your Volkswagen. Even with
>>>>>>>>> big LDMOS parts I had that happen. One company (PolyFet) stuck out and
>>>>>>>>> provided SPICE models. So the big competitor will be left out ... sorry,
>>>>>>>>> NXP.
>>>>>>>> I don't simulate this stuff, I test. Most microwave parts are
>>>>>>>> characterized small-signal, S-params. Spice models are very rare in
>>>>>>>> this business. I know things about some of these parts that the
>>>>>>>> manufacturers don't know.
>>>>>>>>
>>>>>>> Lately I have simulated more, and then straight to layout. The one
>>>>>>> that's in fab right now is a white-knuckle ride because I am using a
>>>>>>> device in a weird way. Usually pans out though.
>>>>>>>
>>>>>>> What often amazes me is how close to abs max people (and sometimes app
>>>>>>> notes) bias RF devices. When it says 7V they bias them to 5V or 6V on
>>>>>>> the collector/drain. Some day a huge pulse finds its way into the input,
>>>>>>> almost saturates the device, then it lets go, the inductor shoots up ...
>>>>>>> phut.
>>>>>> I've run 7-volt-rated MESfets at 18 volts, and 2 volt RF schottkies at
>>>>>> 6. I sometimes test parts to destruction and then back off some. The
>>>>>> performance is often worth the small risk.
>>>>>>
>>>>> I've seen that. But if you are a consultant and this goes into some
>>>>> aircraft you can't design like that. Even with an indemnification clause
>>>>> that can cause lots of grief should it go wrong in only one case. It's
>>>>> got to be by the book.
>>>>>
>>>>>
>>>>>> Lots of RF devices seem to have voltage ratings that assume an RF tank
>>>>>> in the output, so that the actual instantaneous drain excursion is 0
>>>>>> to twice Vcc. The RF guys are really, really terrible when it comes to
>>>>>> realtime specs. You're lucky to get any DC curves. More often there's
>>>>>> an app circuit with an input match, an output match, and a gate bias
>>>>>> trimpot.
>>>>>>
>>>>> All they really offer in many cases is a set of S-parameters. Since I
>>>>> design a lot of pulse circuitry I almost do a rain dance every time
>>>>> there is just a snippet of SPICE data.
>>>> Somewhere in my bag-o-tricks is an S-Parameter-to-Spice-Model
>>>> converter (and vice-versa).
>>>>
>>>> When I find it I'll post.
>>>>
>>> That would be very kind.
>>>
>>> However, except for super large transmitter devices (and sometimes even
>>> then) the S-parameters have been measured under small signal conditions.
>>> Doesn't help much when pulsing them hard, then you are back to what John
>>> said, experiments on the lab bench.
>>
>> The RF boys call that "load pull" testing. People make interfaced
>> load-pull boxes so you can automate exploring the nonlinear space of
>> large-signal RF amps. Between explosions, anyhow.
>>
>
>Mostly I just press on, eye-ball the rough L and C values, go into the
>garage and see if I have enough suitable variable capacitor for a test.
>Then it's off to the plumber's place for some copper tubing, to make the
>coils. At Home Depot that's now called "plomeria".
>
>
>> google "load pull tuner"
>>
>> Got one of those coming up. I am
>>> going to do the first run with tubes again to see if the concept pans
>>> out, in order not to have stuff blow up in my face. It's really nice to
>>> know that one can still buy the mil version of the 6146 tube, at
>>> affordable prices. So far I haven't blown one though, only tired out a pair.
>>
>> My first job interview, I told the guy that I preferred tubes to
>> transistors because tubes were harder to blow up. The smug SOB said
>> "that won't do" and sent me away. My next interview, I said the same
>> thing, and Melvin laughed and hired me. I designed $200 million worth
>> of stuff for him.
>>
>
>I've had a similar event with a potential client a long time ago.
>Proposed an unorthodox solution that seemed to be "too simple" for them,
>not academic enough. They turned me down. A few years later this company
>was at the financial end of the rope and went belly up. The sad thing is
>that my solution would have potentially saved them, along with about two
>dozen jobs :-(


Wonderful simple circuits get no respect in certain quarters.

John

From: Phil Hobbs on
John Larkin wrote:
> On Sat, 12 Sep 2009 13:14:20 -0700, Joerg <invalid(a)invalid.invalid>
> wrote:
>
>> I've had a similar event with a potential client a long time ago.
>> Proposed an unorthodox solution that seemed to be "too simple" for them,
>> not academic enough. They turned me down. A few years later this company
>> was at the financial end of the rope and went belly up. The sad thing is
>> that my solution would have potentially saved them, along with about two
>> dozen jobs :-(
>
>
> Wonderful simple circuits get no respect in certain quarters.
>

That's one definition of "inside the box thinking." It's a shock the
first few times you run into it.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
From: Charlie E. on
On Sat, 12 Sep 2009 10:01:22 -0700, Jim Thompson
<To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> wrote:

>On Sat, 12 Sep 2009 08:51:21 -0700, Joerg <invalid(a)invalid.invalid>
>wrote:
>
>>John Larkin wrote:
>>> On Fri, 11 Sep 2009 16:38:42 -0700, Joerg <invalid(a)invalid.invalid>
>>> wrote:
>>>
>>>> John Larkin wrote:
>>>>> On Fri, 11 Sep 2009 15:33:08 -0700, Joerg <invalid(a)invalid.invalid>
>>>>> wrote:
>>
>>[...]
>>
>>>>
>>>>>> One issue I found with RF devices was that you can sometimes only obtain
>>>>>> small signal simulator models that aren't very helpful with pulse apps,
>>>>>> plus require software that costs more than your Volkswagen. Even with
>>>>>> big LDMOS parts I had that happen. One company (PolyFet) stuck out and
>>>>>> provided SPICE models. So the big competitor will be left out ... sorry,
>>>>>> NXP.
>>>>> I don't simulate this stuff, I test. Most microwave parts are
>>>>> characterized small-signal, S-params. Spice models are very rare in
>>>>> this business. I know things about some of these parts that the
>>>>> manufacturers don't know.
>>>>>
>>>> Lately I have simulated more, and then straight to layout. The one
>>>> that's in fab right now is a white-knuckle ride because I am using a
>>>> device in a weird way. Usually pans out though.
>>>>
>>>> What often amazes me is how close to abs max people (and sometimes app
>>>> notes) bias RF devices. When it says 7V they bias them to 5V or 6V on
>>>> the collector/drain. Some day a huge pulse finds its way into the input,
>>>> almost saturates the device, then it lets go, the inductor shoots up ...
>>>> phut.
>>>
>>> I've run 7-volt-rated MESfets at 18 volts, and 2 volt RF schottkies at
>>> 6. I sometimes test parts to destruction and then back off some. The
>>> performance is often worth the small risk.
>>>
>>
>>I've seen that. But if you are a consultant and this goes into some
>>aircraft you can't design like that. Even with an indemnification clause
>>that can cause lots of grief should it go wrong in only one case. It's
>>got to be by the book.
>>
>>
>>> Lots of RF devices seem to have voltage ratings that assume an RF tank
>>> in the output, so that the actual instantaneous drain excursion is 0
>>> to twice Vcc. The RF guys are really, really terrible when it comes to
>>> realtime specs. You're lucky to get any DC curves. More often there's
>>> an app circuit with an input match, an output match, and a gate bias
>>> trimpot.
>>>
>>
>>All they really offer in many cases is a set of S-parameters. Since I
>>design a lot of pulse circuitry I almost do a rain dance every time
>>there is just a snippet of SPICE data.
>
>Somewhere in my bag-o-tricks is an S-Parameter-to-Spice-Model
>converter (and vice-versa).
>
>When I find it I'll post.
>
> ...Jim Thompson

Yes, Jim, I think I sent that to you back in the day. But, you have
to be careful. It really does what is in effect an inverse FFT of the
Sparams, so it doesn't include any of the really important data you
need to do a good spice simulation when you are concerned more with
transient response. IBIS models are a 'little' better, but they are
also very behavioral and I was always interested in the little
'gotchas' that following them too closely led to.

Charlie
From: Joerg on
Charlie E. wrote:
> On Sat, 12 Sep 2009 10:01:22 -0700, Jim Thompson
> <To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> wrote:
>
>> On Sat, 12 Sep 2009 08:51:21 -0700, Joerg <invalid(a)invalid.invalid>
>> wrote:
>>
>>> John Larkin wrote:
>>>> On Fri, 11 Sep 2009 16:38:42 -0700, Joerg <invalid(a)invalid.invalid>
>>>> wrote:
>>>>
>>>>> John Larkin wrote:
>>>>>> On Fri, 11 Sep 2009 15:33:08 -0700, Joerg <invalid(a)invalid.invalid>
>>>>>> wrote:
>>> [...]
>>>
>>>>>>> One issue I found with RF devices was that you can sometimes only obtain
>>>>>>> small signal simulator models that aren't very helpful with pulse apps,
>>>>>>> plus require software that costs more than your Volkswagen. Even with
>>>>>>> big LDMOS parts I had that happen. One company (PolyFet) stuck out and
>>>>>>> provided SPICE models. So the big competitor will be left out ... sorry,
>>>>>>> NXP.
>>>>>> I don't simulate this stuff, I test. Most microwave parts are
>>>>>> characterized small-signal, S-params. Spice models are very rare in
>>>>>> this business. I know things about some of these parts that the
>>>>>> manufacturers don't know.
>>>>>>
>>>>> Lately I have simulated more, and then straight to layout. The one
>>>>> that's in fab right now is a white-knuckle ride because I am using a
>>>>> device in a weird way. Usually pans out though.
>>>>>
>>>>> What often amazes me is how close to abs max people (and sometimes app
>>>>> notes) bias RF devices. When it says 7V they bias them to 5V or 6V on
>>>>> the collector/drain. Some day a huge pulse finds its way into the input,
>>>>> almost saturates the device, then it lets go, the inductor shoots up ...
>>>>> phut.
>>>> I've run 7-volt-rated MESfets at 18 volts, and 2 volt RF schottkies at
>>>> 6. I sometimes test parts to destruction and then back off some. The
>>>> performance is often worth the small risk.
>>>>
>>> I've seen that. But if you are a consultant and this goes into some
>>> aircraft you can't design like that. Even with an indemnification clause
>>> that can cause lots of grief should it go wrong in only one case. It's
>>> got to be by the book.
>>>
>>>
>>>> Lots of RF devices seem to have voltage ratings that assume an RF tank
>>>> in the output, so that the actual instantaneous drain excursion is 0
>>>> to twice Vcc. The RF guys are really, really terrible when it comes to
>>>> realtime specs. You're lucky to get any DC curves. More often there's
>>>> an app circuit with an input match, an output match, and a gate bias
>>>> trimpot.
>>>>
>>> All they really offer in many cases is a set of S-parameters. Since I
>>> design a lot of pulse circuitry I almost do a rain dance every time
>>> there is just a snippet of SPICE data.
>> Somewhere in my bag-o-tricks is an S-Parameter-to-Spice-Model
>> converter (and vice-versa).
>>
>> When I find it I'll post.
>>
>> ...Jim Thompson
>
> Yes, Jim, I think I sent that to you back in the day. But, you have
> to be careful. It really does what is in effect an inverse FFT of the
> Sparams, so it doesn't include any of the really important data you
> need to do a good spice simulation when you are concerned more with
> transient response.


That's exactly the point, especially with LDMOS. When a big old fat one
goes phut ... *KABLAM* it's mucho Dolares down the drain. Not the best
way to make friends at a new client.


> ... IBIS models are a 'little' better, but they are
> also very behavioral and I was always interested in the little
> 'gotchas' that following them too closely led to.
>
> Charlie


--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.