GR1862-B help, please
in [Design]
|
From: Robert Baer on 23 Apr 2010 22:57 ...not a happy camper re my GR. Those two 100Meg resistors inside were rather high; one about 9% high and the other about 4% high - a good reason why the 100Gohm 1% resistor i bought (as a reference) read around 5% high on the GR. So..i bought two SlimMox 100Meg 1% 5KV rating resistors for replacement; those seem to be roughly 0.2% high and very close to each other in value (using a Mastech MY65 and a 100Meg 0.1% resistor as a reference). Used teflon sleeving for the heck of it, even tho the leads are in air after mounting. And the values of the replacement resistors read essentially the same afterwards. *BUT* That 100Gohm now reads higher..about 15% high(!!) on the GR. Using the 500V from the GR and my new Mastech as a (10Meg input Z) current meter (200nA FS), i read 5.046nA; with 498.7V supply giving a calculated 98.8Kmegs which is within the 1% spec or OK. So thar bee sumppin rong widde GR! Any crazy ideas?
From: Robert Baer on 24 Apr 2010 19:19 Robert Baer wrote: > ..not a happy camper re my GR. > Those two 100Meg resistors inside were rather high; one about 9% high > and the other about 4% high - a good reason why the 100Gohm 1% resistor > i bought (as a reference) read around 5% high on the GR. > So..i bought two SlimMox 100Meg 1% 5KV rating resistors for > replacement; those seem to be roughly 0.2% high and very close to each > other in value (using a Mastech MY65 and a 100Meg 0.1% resistor as a > reference). > Used teflon sleeving for the heck of it, even tho the leads are in air > after mounting. > And the values of the replacement resistors read essentially the same > afterwards. > *BUT* > That 100Gohm now reads higher..about 15% high(!!) on the GR. > Using the 500V from the GR and my new Mastech as a (10Meg input Z) > current meter (200nA FS), i read 5.046nA; with 498.7V supply giving a > calculated 98.8Kmegs which is within the 1% spec or OK. > > So thar bee sumppin rong widde GR! > Any crazy ideas? Well, i got nasty and used LTspice and simulated the beast; found that 100pA of grid current can give/add 2% error. So...since i have already made fixes (changed a 600V cap that leaked more as temp went up, changed the two 100Meg resistors), i am now thinking of replacing the 12AU7 with a dual JFET. Granted i will have to lower the "plate" voltage from the existing 45V to something in the region of 15V. But i do not know what JFETs would do as the worst case gate current at (say) 50C (as a wild guess of warm-up temp inside the GR) needs to be 10pA or better. Any suggestions?
From: Joerg on 24 Apr 2010 20:56 Robert Baer wrote: > Robert Baer wrote: >> ..not a happy camper re my GR. >> Those two 100Meg resistors inside were rather high; one about 9% >> high and the other about 4% high - a good reason why the 100Gohm 1% >> resistor i bought (as a reference) read around 5% high on the GR. >> So..i bought two SlimMox 100Meg 1% 5KV rating resistors for >> replacement; those seem to be roughly 0.2% high and very close to each >> other in value (using a Mastech MY65 and a 100Meg 0.1% resistor as a >> reference). >> Used teflon sleeving for the heck of it, even tho the leads are in >> air after mounting. >> And the values of the replacement resistors read essentially the >> same afterwards. >> *BUT* >> That 100Gohm now reads higher..about 15% high(!!) on the GR. >> Using the 500V from the GR and my new Mastech as a (10Meg input Z) >> current meter (200nA FS), i read 5.046nA; with 498.7V supply giving a >> calculated 98.8Kmegs which is within the 1% spec or OK. >> >> So thar bee sumppin rong widde GR! >> Any crazy ideas? > Well, i got nasty and used LTspice and simulated the beast; found that > 100pA of grid current can give/add 2% error. > So...since i have already made fixes (changed a 600V cap that leaked > more as temp went up, changed the two 100Meg resistors), i am now > thinking of replacing the 12AU7 with a dual JFET. Here goes another classic piece of electronics ... > Granted i will have to lower the "plate" voltage from the existing 45V > to something in the region of 15V. > But i do not know what JFETs would do as the worst case gate current > at (say) 50C (as a wild guess of warm-up temp inside the GR) needs to be > 10pA or better. > Any suggestions? If you can get your hands on one of these, they are supposed to be very low leakage: http://www.fairchildsemi.com/ds/PN/PN4117A.pdf Plus you might want to move to a place with zero humidity and no smoking within 250ft of unit :-) -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
From: Robert Baer on 26 Apr 2010 03:38 Joerg wrote: > Robert Baer wrote: >> Robert Baer wrote: >>> ..not a happy camper re my GR. >>> Those two 100Meg resistors inside were rather high; one about 9% >>> high and the other about 4% high - a good reason why the 100Gohm 1% >>> resistor i bought (as a reference) read around 5% high on the GR. >>> So..i bought two SlimMox 100Meg 1% 5KV rating resistors for >>> replacement; those seem to be roughly 0.2% high and very close to >>> each other in value (using a Mastech MY65 and a 100Meg 0.1% resistor >>> as a reference). >>> Used teflon sleeving for the heck of it, even tho the leads are in >>> air after mounting. >>> And the values of the replacement resistors read essentially the >>> same afterwards. >>> *BUT* >>> That 100Gohm now reads higher..about 15% high(!!) on the GR. >>> Using the 500V from the GR and my new Mastech as a (10Meg input Z) >>> current meter (200nA FS), i read 5.046nA; with 498.7V supply giving a >>> calculated 98.8Kmegs which is within the 1% spec or OK. >>> >>> So thar bee sumppin rong widde GR! >>> Any crazy ideas? >> Well, i got nasty and used LTspice and simulated the beast; found >> that 100pA of grid current can give/add 2% error. >> So...since i have already made fixes (changed a 600V cap that leaked >> more as temp went up, changed the two 100Meg resistors), i am now >> thinking of replacing the 12AU7 with a dual JFET. > > > Here goes another classic piece of electronics ... > > >> Granted i will have to lower the "plate" voltage from the existing >> 45V to something in the region of 15V. >> But i do not know what JFETs would do as the worst case gate current >> at (say) 50C (as a wild guess of warm-up temp inside the GR) needs to >> be 10pA or better. >> Any suggestions? > > > If you can get your hands on one of these, they are supposed to be very > low leakage: > > http://www.fairchildsemi.com/ds/PN/PN4117A.pdf > > Plus you might want to move to a place with zero humidity and no smoking > within 250ft of unit :-) > Thanks for the reference! I have never smoked, so that is not a problem; also there is no characteristic smoke smell in the unit. For low / zero humidity, ARIDzona would seem to be the place.
From: Joerg on 26 Apr 2010 12:49
Robert Baer wrote: > Joerg wrote: >> Robert Baer wrote: >>> Robert Baer wrote: >>>> ..not a happy camper re my GR. >>>> Those two 100Meg resistors inside were rather high; one about 9% >>>> high and the other about 4% high - a good reason why the 100Gohm 1% >>>> resistor i bought (as a reference) read around 5% high on the GR. >>>> So..i bought two SlimMox 100Meg 1% 5KV rating resistors for >>>> replacement; those seem to be roughly 0.2% high and very close to >>>> each other in value (using a Mastech MY65 and a 100Meg 0.1% resistor >>>> as a reference). >>>> Used teflon sleeving for the heck of it, even tho the leads are in >>>> air after mounting. >>>> And the values of the replacement resistors read essentially the >>>> same afterwards. >>>> *BUT* >>>> That 100Gohm now reads higher..about 15% high(!!) on the GR. >>>> Using the 500V from the GR and my new Mastech as a (10Meg input Z) >>>> current meter (200nA FS), i read 5.046nA; with 498.7V supply giving >>>> a calculated 98.8Kmegs which is within the 1% spec or OK. >>>> >>>> So thar bee sumppin rong widde GR! >>>> Any crazy ideas? >>> Well, i got nasty and used LTspice and simulated the beast; found >>> that 100pA of grid current can give/add 2% error. >>> So...since i have already made fixes (changed a 600V cap that >>> leaked more as temp went up, changed the two 100Meg resistors), i am >>> now thinking of replacing the 12AU7 with a dual JFET. >> >> >> Here goes another classic piece of electronics ... >> >> >>> Granted i will have to lower the "plate" voltage from the existing >>> 45V to something in the region of 15V. >>> But i do not know what JFETs would do as the worst case gate >>> current at (say) 50C (as a wild guess of warm-up temp inside the GR) >>> needs to be 10pA or better. >>> Any suggestions? >> >> >> If you can get your hands on one of these, they are supposed to be >> very low leakage: >> >> http://www.fairchildsemi.com/ds/PN/PN4117A.pdf >> >> Plus you might want to move to a place with zero humidity and no >> smoking within 250ft of unit :-) >> > Thanks for the reference! > I have never smoked, so that is not a problem; also there is no > characteristic smoke smell in the unit. > For low / zero humidity, ARIDzona would seem to be the place. Yes, Ari-"but its a dry heat"-zona would be the place. Although, when I think about Jim's problem with gnats I am not so sure about the situation right now :-) -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM. |