Thank you Pete. I think I'll wait for the boards to get here and see if I can squeeze in a 1.5uF/450V Panasonic ECQ-E2W155KC
Thank you Pete. I think I'll wait for the boards to get here and see if I can squeeze in a 1.5uF/450V Panasonic ECQ-E2W155KC
Board arrived
The above would be a very tight fit, but it might just fit. However, ECQ-E2W105KC will definitely fit and is a 1.0uF/450V part.
I have some insight here too. We used to have a semiconductor division until they were spun off, and then all of the local people were laid off, and then the offices were closed. The same thing happened at other semiconductor houses. I have friends that were laid off by Atmel, Freescale, On Semi, Linear Technology and National Semi. Many of the independent sales reps dissapeared too. The building that I work in designs cellular phones. Our production volumes are about 1/4 what they were two years ago. The design teams are much smaller than they were two years ago, and a lot of work is ODM'ed. The IC sales reps that used to visit once or twice a week, don't come around any more, and some won't even return your calls.
As world wide production of phones, PC's and other technology intensive products plummeted a year ago many IC companies had major layoffs, and even closed down some wafer fabs. This was especially true of IC products on old fabs, weren't profitable, or were only set up and run periodically. Some of these chips will never return at some companies. That just makes the waiting list longer at the places making chips, snd he who buys the biggest volume can cut in line!
We, who used to piggyback on the volumes of the cellular division to build products with comparatively tiny volumes can't get parts either.
I work for a large German company with diverse divisions and production world wide. They used to have a cell phone division, but sold it off years ago. Same with our semiconductor divisions. I'm in Industrial Controls, so we are considered low volume as a lot of the parts we use have demand for less than 100,000 AOQ.
I can still occasionally get free samples though
I can still occasionally get free samples though
Pete:
I believe these files need updating as well:
http://www.pmillett.com/file_downloads/ATEST.pdf
http://www.pmillett.com/file_downloads/ATEST_BOM.pdf
$%@# Microsoft FrontPage!
I may have to manually delete the files from the server to get them to update. For some reason FP did not upload the new versions, even though they are updated in my local copy.
I have been fighting the same thing.
FP went extinct a few years ago. MS released the last version in 2003, but kept supporting the Front Page extensions (even on Linux hosting servers) until last September. Many hosting companies just turned the extensions off at that time. They are required for my web site to work correctly, and my hosting company is still using them with no tech support.
My web site was built in FP and I let FP handle the entire navigation structure. This has now led to the same thing you are finding. Old pages do not update, and new pages do not even show up in the web site even though the files were transfered to the hosting server.
I find that the only way I can guarantee that the remote site will match my local copy is to manually delete the entire web site from the host server, then let FP upload everything again.
I guess that it is time to move beyond FP. I have chosed MS Expression Web. I hate to go down the MS route again, but it was cheap ($70).
OK, I think they are updated now. I manually deleted them and re-published. You might have to ctrl-refresh to clear the cached copy...
Pete
Got my board and am slowly stuffing it as budget (tiny) permits. May be some time before I implement the meter display parts.
Just for fun, I ordered a sheet of Ultraperm magnetic shielding material:
Rare Ultraperm 80 Shielding Sheet-The Electronic Goldmine
Been wanting to get some ERS 'paper' for RFI; not sure it will be of much application here; I think the real need is for low frequency (60 Hz) magnetic shielding.
Surplus Sales of Nebraska had nearly all the resistors I wanted in RN55E (25 ppm) or equivalent parts. They've got some high precision Vishay foil 50K resistors, which might be good for the input section; these are probably the lowest noise resistors available. I just don't like the idea of using these parts as possible fuses.
My original thought was to use a AC VTVM such as the Heathkit IM-5238 as a front end for a sound card analyzer, but it's just too noisy and distorted to use without major mods. With the Millett Meter (tm), I may have that budget Audio Precision test set I could never hope to afford...
Just for fun, I ordered a sheet of Ultraperm magnetic shielding material:
Rare Ultraperm 80 Shielding Sheet-The Electronic Goldmine
Been wanting to get some ERS 'paper' for RFI; not sure it will be of much application here; I think the real need is for low frequency (60 Hz) magnetic shielding.
Surplus Sales of Nebraska had nearly all the resistors I wanted in RN55E (25 ppm) or equivalent parts. They've got some high precision Vishay foil 50K resistors, which might be good for the input section; these are probably the lowest noise resistors available. I just don't like the idea of using these parts as possible fuses.
My original thought was to use a AC VTVM such as the Heathkit IM-5238 as a front end for a sound card analyzer, but it's just too noisy and distorted to use without major mods. With the Millett Meter (tm)
, I may have that budget Audio Precision test set I could never hope to afford...
Just be careful with the shielding material, it changes magnetic properties (looses permeability) when bent or deformed. It is normally bent to shape, THEN annealed to remove stresses that reduce the permeability. Cut carefully.
Once you have it installed, run some tests with the enclosure assembled, and with the top off and let us know if it makes much difference in the 50-200hz range.
Once you have it installed, run some tests with the enclosure assembled, and with the top off and let us know if it makes much difference in the 50-200hz range.
It might be worthwhile to add / mention four #4-40 x 0.25" sheet metal screws for securing the board to the bottom of the case. Of course, with all the panel mount components, perhaps it isn't a huge deal. Stuff like this isn't easy to get at Mouser, but they sell them, as long as you don't mind a few extras ( 5721-4-1/4-SS).
Can anyone tell me what the normal current consumption of the converter happens to be? I had to substitute a different PTC (polyfuse) for the one Pete recommended and with all of the ICs installed the PTC is tripping and shutting the interface off. (The input and output voltage drops like a rock some seconds after power is applied.) The PTC is rated for 500mA continuous, but perhaps I missed something in its specification. Unloaded interface with no signals present I would not have expected the current draw from the converter to be over 150mA or so. (Naive? Guess I will go back and read the spec.)
With just a couple of ICs installed (the input chain) I measured 250mA of current drawn on the 5V supply which seems very high to me. No signs of obvious problems either - voltages seem normal and it will pass clean signals. I notice with no cable connected to the sound card "input" jack that the converter shuts down immediately which leaves me wondering if the ThatCorp 1646 driver IC might be oscillating or that the outputs might be shorted in this mode causing high current draw. Installing any of the ICs on the generator side causes this problem.
I will cobble up an external +/-15V supply at some point and see what is going on.
This converter is pretty noisy too, I noticed spikes on the dc output of more than several hundred mV (it was late so I do not recall the period, but I believe it was few uS or so)
The supply unloaded produces +/-15V.. I think it is actually working OK and there is something that is loading down the supply excessively. (Oscillating?)
I did not install the meter circuitry, and the additional bypass caps those ICs used so I may try some additional rail capacitance and possibly ceramic bypass caps on some of the ICs. All components have been carefully checked against the BOM and are correct. (As I indicated parts of the unit seem to work under some conditions.)
I will continue to investigate what I have done wrong, but so far nothing is leaping out at me.
With just a couple of ICs installed (the input chain) I measured 250mA of current drawn on the 5V supply which seems very high to me. No signs of obvious problems either - voltages seem normal and it will pass clean signals. I notice with no cable connected to the sound card "input" jack that the converter shuts down immediately which leaves me wondering if the ThatCorp 1646 driver IC might be oscillating or that the outputs might be shorted in this mode causing high current draw. Installing any of the ICs on the generator side causes this problem.
I will cobble up an external +/-15V supply at some point and see what is going on.
This converter is pretty noisy too, I noticed spikes on the dc output of more than several hundred mV (it was late so I do not recall the period, but I believe it was few uS or so)
The supply unloaded produces +/-15V.. I think it is actually working OK and there is something that is loading down the supply excessively. (Oscillating?)
I did not install the meter circuitry, and the additional bypass caps those ICs used so I may try some additional rail capacitance and possibly ceramic bypass caps on some of the ICs. All components have been carefully checked against the BOM and are correct. (As I indicated parts of the unit seem to work under some conditions.)
I will continue to investigate what I have done wrong, but so far nothing is leaping out at me.