SUPER-DUPER PIN DIODE needed - RF Cafe Forums
Because of the high maintenance needed to monitor and filter spammers from the RF Cafe Forums, I decided that it would be best to just archive the pages to make all the good information posted in the past available for review. It is unfortunate that the scumbags of the world ruin an otherwise useful venue for people wanting to exchanged useful ideas and views. It seems that the more formal social media like Facebook pretty much dominate this kind of venue anymore anyway, so if you would like to post something on RF Cafe's Facebook page, please do.
Below are all of the forum threads, including all the responses to the original posts.
Post subject: SUPER-DUPER PIN DIODE needed Posted: Mon Oct 17, 2005 5:58 am
I never used pin diodes in my projects, hence looking for your, guys, advice.
The project is a magnetic loop autotune antenna. Currently it is tuned by vacuum cap driven by stepper and works fine with average tuning speed of 3-4 seconds. On lower bands It needs additional caps and these added by jennings vaccum relay that also controlled by the same CPU. These caps and relays are 6kV/20Amps rated, lower rated arching, checked. The loop itself is a very high Q circuit and has reactive impedance. Although 100 watt PEP is the power I have to multiply it by Q=900 in order to rate these components sufficiently.
Now I want to use my FH (freq. hooping) radio with my antenna and it requires 4-5 millisec tuning time for the antenna. This means that I have to switch from motor tuning to SPST relay switcheable capacitor bank. However, all suitable relays have 2.5-3ms switching time (it is measured, all of them rated 10ms) which means there is almost no room left for calculations. These relays have 10-12 mOhms in closed contacts which is next to bad as larger resistance causes antenna radiating resistance less than sum of other resistances and lowers Q and efficiency dramatically.
The question is:
Is the nowdays pin diode technology advanced enough to offer following parameters: (reverse voltage 10-15kV, open diode resistance 0.01 Ohms, Current 20-30 Amps (at 30 MHz)). Are they available commercially? In what packages? (SMD is preffered).
I think switching speed is out of qustion, all of them are fast enough.
Browsing on pin diode webs I noted that most powerful an low resistance (best I found is 0.1 Ohm) are biased with quite high current. I have only 3 Amps for 16 diodes.
Thanks and regards,
Post subject: Super-duper PIN diodePosted: Mon Oct 17, 2005 10:52 pm
I don't think you'll find what you're looking for.
To the best of my knowledge, there are no SMT diodes of any variety rated at 10-15 kV. One reason is that we're talking electric field densities near or over the breakdown of air at STP for most SMT packages.
Secondly, look at the values you name for current and resistance. Even though you can't run a PIN diode at zero forward current, taking your low end estimate of 20 Amps and 0.01 Ohms on resistance, P = I squared R gives 4 Watts of power dissipation: rather a lot for an SMT.
Finally, you have to have sufficient carriers in the intrinsic region to sustain the currents you need. As a rule of thumb, that means a forward ON-state current of more than 20 Amps - and you only have 3/16 A!
On the bright side, 1 millisecond for a calculation, even on a cheap uP, seems like a long time!
Post subject: Posted: Tue Oct 18, 2005 3:00 am
Thanks for that.
look at http://www.microsemi.com/datasheets/hum3002-04 .pdf
These ones, if three or four in parallel are would be a subject for experiment (20-30 mOhms) if were not packaged in these railways styled bolts. Also, it makes me optimistic that since these already come quite close to my specs I can hope on soon availability of better ones.
However reverse volatge is still too low. Currently I'am switching 7kV/12A rated multilayer ceramic caps by 6kV/12A/0.012Ohms relays. These caps in most cases added to vaccum cap at nearly full capacitance (typical situation) in parallel, thus current is divided between them and they getting warm very little.
The nature of entire circuit, as you can see, does not allow for conducting of precise measurements. The q-factor is derived as calculation from measurement of bandwidth at resonance curve. Thus, though powered by 100 Watt PEP the reactive power has to be considered as much as 90 kW. Lots of guessing.
The fact that other guys managed to do similar job with 1 kW (!) powered transmitter makes me very intrigued. By what means??? All known relays are too slow, reed relays too weak and have huge capacitance to it's coil... Magic.
Thanks and regards,
Post subject: Super-duper PI diodesPosted: Wed Oct 19, 2005 1:52 am
Regarding those Microsemi PIN diodes: wow!
Thanks for bringing those to my attention. Maybe your task isn't so impossible after all. But what's the problem with the package? They obviously need to get rid of about 50W of heat.
Post subject: Posted: Wed Oct 19, 2005 5:29 am
In my case I do not face trouble of heat dissipation this much. First of all I do not have CW all the time, SSB heats the diode by carrier that appears during modulation only, pauses are for cooling. The other thing is typical fashion of communications is 10:1, e.g. you listen more than speak. For that reason conventional air cooling would be sufficient. Also construction, having SMD enveloped multilayer caps and chip like pin diodes placed over copper clad flan (I think FR-4 also will do, however) makes whole thing much more organized, compact and technologically easier for assembling. With these bolts and strip lided caps entire construction will look like 1930s SX-28 shassis.
Post subject: Posted: Wed Oct 19, 2005 5:43 am
....and again, they need their bolts exactly because 0.1Ohms resistance, should they have made three same crystals on common piece of waffer and packaged into ceramic or plastic cube shaped SMD, they would get 3 times less lossy unit with no heatsink requirements for 50W for sure. Just rerate by frequency, not all use them for microwave...I think this package is bit overengineered.
More than 10,000 searchable pages indexed.