jackinnj said:
Now if you manage to get something to -320oC you are up for a Nobel prize in physics..... because absolute zero is -273.15oC! Or 0 Kelvin.
Or you were thinking of those strange Fahrenheit we never use over here.
Secondly yes, LH2 (at 4K, or was that LHe) would be perfect. That would even make several metals super-conductors
! But LN2 is in theory available to normal people, but LH2 is NOT. Several OCR's use LN2 as it is in fact really easy to get. Hey I even know how to make it myself
. I'll just make a Stirling cryocooler. But I think I'll try to finish the AlephX' first 
And if I'll ever build a stirling cryocooler I'll use it to make superconducting magnet system for a compressiondriver for sub-bass! Anyone know how to make an amp capable of driving 0 ohm (as in super-conducting voicecoil...)?
Ups.... sorry for being OT, just couldn't resist.
Anders
akb1212 said:
MEA CULPA, MEA MAXIMA CULPA !!!
I actually worked with the stuff in college. There was a jesuit priest who ran the lab -- occasionally he would take a hot dog, dip it into the nitrogen dewar, put it on the lab bench and whack it with a mallet!
The LN was just the pre-cooler, hydrogen was used to get it close to absoute zero and we would watch the conductivity as a function of temperature and magnetic field.
My claim to fame was building a phase-lock loop amplifier with tubes!
Sencore TF46
I have been matching FETs and BJTs for more than 10 years using a multimeter and a DC voltage source, as I am seeing on this thread.
Q1: In the book, the Sencore TF46 is recommended for matching purposes. This tool sells for about $100 on ebay. I have never seen one and don't know anyone who has used it.
Simply put, what is its value ? what does it do better than a typical lab set-up for matching ?
Q2: I have seen in one article that frequency matching should be done. I never did this and do not intend on doing it unless someone exposes a real good reason. But rather, should matching be done at an arbitrary frequency of say 1kHz ?
I have been matching FETs and BJTs for more than 10 years using a multimeter and a DC voltage source, as I am seeing on this thread.
Q1: In the book, the Sencore TF46 is recommended for matching purposes. This tool sells for about $100 on ebay. I have never seen one and don't know anyone who has used it.
Simply put, what is its value ? what does it do better than a typical lab set-up for matching ?
Q2: I have seen in one article that frequency matching should be done. I never did this and do not intend on doing it unless someone exposes a real good reason. But rather, should matching be done at an arbitrary frequency of say 1kHz ?
Re: Sencore TF46
if you have to match for frequency response (small signal) you might be better off using one of the transistor matrices from Intersil -- (kind of a legacy business from when it was Harris) in fact, when I look at the discrete crossover group-buy I wonder whether they are aware of the consequences of limiting the bandwidth in a discrete op-amp.
i do know that the discrete transistor comparator I built for my phase meter didn't work as well as the Intersil component where all devices were on one die --
Demir said:
if you have to match for frequency response (small signal) you might be better off using one of the transistor matrices from Intersil -- (kind of a legacy business from when it was Harris) in fact, when I look at the discrete crossover group-buy I wonder whether they are aware of the consequences of limiting the bandwidth in a discrete op-amp.
i do know that the discrete transistor comparator I built for my phase meter didn't work as well as the Intersil component where all devices were on one die --
How far off are chips off the same wafer?
Just out of curiousity, is the matching of IRF-510s for use in the BOZ critical?
I'm not sure since these aren't mirrored against each other in a balanced configuration.
In this case, is it just a case of matching gain?
Finally, just how far off in this regard are chips from the same lot? If they tend to be pretty close, I might forgo the matching if it's not that critical for the BOZ.
Thanks,
KT
Just out of curiousity, is the matching of IRF-510s for use in the BOZ critical?
I'm not sure since these aren't mirrored against each other in a balanced configuration.
In this case, is it just a case of matching gain?
Finally, just how far off in this regard are chips from the same lot? If they tend to be pretty close, I might forgo the matching if it's not that critical for the BOZ.
Thanks,
KT
Can I match mosfet device with an UNREGULATE power supply. The output is 33vdc, only a transformer, bridge, and a 330uf cap.
And if I can do my test with this equipment, what is the value of the resistor I need to use for the IRFP044 match ???
I read the article of NP but I'm not very sure for the otput device.
And if I can do my test with this equipment, what is the value of the resistor I need to use for the IRFP044 match ???
I read the article of NP but I'm not very sure for the otput device.
how many fets are you matching? i'd only use a regulated supply, I used a 11 ohm resistor 20 watt for that job on the irfp044s, i forget what voltage, but .75A and 30 sec. ...any more time and they get too hot to hold good luck on that
btw, i have some matched groups of 4 on ebay
search for "irfp*"
later
-Matthew K. Olson
btw, i have some matched groups of 4 on ebay
search for "irfp*"
later
-Matthew K. Olson
do you need 4 groups of 3 fets for that? you may not get the close matching you are looking for with only a group of 20. just fyi
here's my ebay link, they are groups of 4, but these'll work for ya
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=4666&item=3868549813&rd=1&ssPageName=WDVW
anyway, if you still decide to match them yourself, in order to get them within desired tolerances, definitely use a regulated supply, and you may need 10-20 more devices to get good matched sets, good luck!
-Matthew K. Olson
here's my ebay link, they are groups of 4, but these'll work for ya
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=4666&item=3868549813&rd=1&ssPageName=WDVW
anyway, if you still decide to match them yourself, in order to get them within desired tolerances, definitely use a regulated supply, and you may need 10-20 more devices to get good matched sets, good luck!
-Matthew K. Olson
arold19,
I would take Matthew up on his ebay offer. You need a well regulated power supply because you will be making DC voltage readings down to the tens of millivolts. Without a regulated supply it will be difficult to get a stable reading. I have used Matts matched fets with great success.
I would take Matthew up on his ebay offer. You need a well regulated power supply because you will be making DC voltage readings down to the tens of millivolts. Without a regulated supply it will be difficult to get a stable reading. I have used Matts matched fets with great success.
OK, I think I will have a problem ! My order is done to digikey, I will receive my peace monday.
I have 20 device in my order, I hope this will be enough ? I will make all the measure and I will post my result and we will decide if the match is correct.
But with a unregulate power supply, with less that 200mA, I don't think it will make ripple ??? I have a scope, I can try to look de DC if it is correct. If you think it is impossible fo match my mosfet with an unregulated power supply, I will make my test at my job.
Thank you very much.
I have 20 device in my order, I hope this will be enough ? I will make all the measure and I will post my result and we will decide if the match is correct.
But with a unregulate power supply, with less that 200mA, I don't think it will make ripple ??? I have a scope, I can try to look de DC if it is correct. If you think it is impossible fo match my mosfet with an unregulated power supply, I will make my test at my job.
Thank you very much.
Ok, like i've said before, NO... 20 devices will probably not be enough even if you did have a regulated supply, and the ripple has nothing to do with the fact that an unregulated supply will NOT work for this task. You need to precisely regulate voltage over time, and line voltage from house power is not stable minute to minute. Ok, so, if you want to match them yourself ....
#1 buy more FETS
#2 get a regulated power supply
otherwise the tolerances will not likely be tight enough, and the matching will not be accurate
-Matthew K. Olson
#1 buy more FETS
#2 get a regulated power supply
otherwise the tolerances will not likely be tight enough, and the matching will not be accurate
-Matthew K. Olson
arold19 said:
I usually match Vgs to +/- 10 mV -- here are some things you have to be careful of:
1) match all the devices at the same time -- the ambient temperature of the room you are working will affect the results.
2) match the devices for the Vgs at a specific point in time -- (I should set up a micro-controller to do this, I now just watch the darkroom clock) -- I find that after 10 seconds the readings settle down.
3) try not to handle the devices too much -- your fingers will warm them up to body temperature, i.e. above ambient -- and use a anti-static strap (particularly in the winter when humidity is low).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.