Not sure why you would bother building umbilical cable when you can buy it by the foot at McMaster-Carr. Looks like you want seven conductors. 16ga should be plenty. Going through the filters gets me to part number 9936K28 which is $2.72/foot.
Best connector? How about no connector. Get a couple of circular strain reliefs (Skintops or the like), some screw-type terminal blocks, a few crimp (or solder) lugs and you're in business. How often does anyone unhook umbilicals anyway? This arrangement will still allow you to remove the cables without desoldering anything (you will have to get in the chassis, though).
Best connector? How about no connector. Get a couple of circular strain reliefs (Skintops or the like), some screw-type terminal blocks, a few crimp (or solder) lugs and you're in business. How often does anyone unhook umbilicals anyway? This arrangement will still allow you to remove the cables without desoldering anything (you will have to get in the chassis, though).
Best connector? How about no connector.>>
Good point!! But you may want to seperate them easily so, for instance, you can work on the power supply alone on the bench. Nothing to stop you only connecting one end. Cheaper in plugs.
But even so, if you have PSU capacitors in the chassis as well as in the PSU unit, you could still get a shock off the male connector that connects to the female feed from the PSU unit. Only way round that is touchproof connectors like the 8 way Speakon. There are fixes, like a resister to earth where the umbilical enters the chassis to drain any HT away fast, but it's still a potential issue.
Good point!! But you may want to seperate them easily so, for instance, you can work on the power supply alone on the bench. Nothing to stop you only connecting one end. Cheaper in plugs.
But even so, if you have PSU capacitors in the chassis as well as in the PSU unit, you could still get a shock off the male connector that connects to the female feed from the PSU unit. Only way round that is touchproof connectors like the 8 way Speakon. There are fixes, like a resister to earth where the umbilical enters the chassis to drain any HT away fast, but it's still a potential issue.
Two points:
1) in my work I use high voltage cables, up to 5000 V, and they are coaxial RG59, with special HV connectors, (they look like BNC, but they aren't). But there is almost no current involved.
2) If a cable for HV is used, is it worth to put a decoupling RC filter (let's say 1 ohm/10 uF) with the cap directly on the B+ of the OPT ?
Best Regards,
Davide
1) in my work I use high voltage cables, up to 5000 V, and they are coaxial RG59, with special HV connectors, (they look like BNC, but they aren't). But there is almost no current involved.
2) If a cable for HV is used, is it worth to put a decoupling RC filter (let's say 1 ohm/10 uF) with the cap directly on the B+ of the OPT ?
Best Regards,
Davide
I am going to try this next week. My test plan is to build a 2 mt cable, power the AC heaters and attach the scope on the HV wires. If I find 50 Hz, something more tricky has to be done.
One approach can be to twist the AC wires and shield them (Morgan Jones recommend something like this. Then you can put everything in the same shrinking tube.
Another thing could be coaxial for the HV. In radiation detectors we have to run long cables for HV, in industrial environment (you do not want to be close to that things) and this is what we use. The pulses from detectors are much smaller than audio signals.
Another approach could be to go for regulated DC for the heaters. I think rectified and filtered would not change your life if there is cross talk.
I'll let you know how it goes.
Best Regards,
Davide
One approach can be to twist the AC wires and shield them (Morgan Jones recommend something like this. Then you can put everything in the same shrinking tube.
Another thing could be coaxial for the HV. In radiation detectors we have to run long cables for HV, in industrial environment (you do not want to be close to that things) and this is what we use. The pulses from detectors are much smaller than audio signals.
Another approach could be to go for regulated DC for the heaters. I think rectified and filtered would not change your life if there is cross talk.
I'll let you know how it goes.
Best Regards,
Davide
Thanks.
DC for heaters is out of the question for me, I prefer the AC heaters.
Looking forward to your results.
Absolutely no doubt for me. I agree with kevinkr. I'd use the AMP TYCO plugs. We use these in all sorts of industrial, military, and automotive environments. They are rugged, reliable, and safe.
IMHO, having separate chassis without connectors on the cords is inconvenient, unless you perhaps bolt the two chassis together.
IMHO, having separate chassis without connectors on the cords is inconvenient, unless you perhaps bolt the two chassis together.
Yes, provided the insulation of the individual wires in the cord is rated at the proper voltage.
These are the connectors for HV I was speaking about:
SHV connector - Wikipedia, the free encyclopedia
But I checked the prices and they are otragious !
I also check the insulation of the normal coaxial rg58 and rg59, and they have 1900V and 2900V max voltage. So I think it's the way to go.
D.
SHV connector - Wikipedia, the free encyclopedia
But I checked the prices and they are otragious !
I also check the insulation of the normal coaxial rg58 and rg59, and they have 1900V and 2900V max voltage. So I think it's the way to go.
D.
Thanks Joshua, but I am not particularly worried about the insulation, am more worried about noise from the AC heater/filament going into my B+.
I've used the Molex 7.50mm Sabre connectors in the past. They are rated for 600V @ 18A and up to 1000V at lower currents. Insulators are rated to 5kV, IIRC. I always use a safety interlock with HV umbilicals. It's basically a ground-return back from the amp that is required to actuate the main relay that powers the entire supply. With no amp attached, it's not possible for the PSU to turn on. The back of my rack also had a grounded steel shroud protecting the umbilicals from fingers/cats/etc.
Last edited:
So, I attached 2m of straight cables, two for AC 6.3 V and two for B+ and gnd between PSU and the Tubelab simpleSE. I needed to have a load.
I could not detect with the scope any 50 Hz both in the B+ or in the speaker output. note that the AC cables were not even twisted. So I think that with twisted cables for AC and a coaxial for B+ there should not be any problem of cross talking. I would not really use the shielding of the B+ for GND, as this should be the thicker wire of all.
A small RC filter in the amp would not hurt.
Personally I am not sure that I want to go with the separate psu way. With the psu I have, that has very good components, I can use at least four amps, just playing with the rectifier, so I would save a lot of money, but I am not sure is very practical.
Here in Japan transformers are good, but fu***** expensive !!
BR,
Davide
I could not detect with the scope any 50 Hz both in the B+ or in the speaker output. note that the AC cables were not even twisted. So I think that with twisted cables for AC and a coaxial for B+ there should not be any problem of cross talking. I would not really use the shielding of the B+ for GND, as this should be the thicker wire of all.
A small RC filter in the amp would not hurt.
Personally I am not sure that I want to go with the separate psu way. With the psu I have, that has very good components, I can use at least four amps, just playing with the rectifier, so I would save a lot of money, but I am not sure is very practical.
Here in Japan transformers are good, but fu***** expensive !!
BR,
Davide
Im using these connectors there rated at 600v from Mouser.
http://www.switchcraft.com/products/pdf_files/multi.pdf
I have a 6 pin to handle the heaters with twisted 18awg wire and a 4 pin to handle 2 300 volt lines, audio ground, chassis ground. I used cable rated for the 300 volt and it's also has shielded that I have tied on one side to chassis ground.
I only did a separate power supply because of weight, The amp would be over a 100lbs if I had it all in one chassis. but the separate power supply made it a lot more expensive and a lot more work. I'm going to think twice before I ever do it again.
http://www.switchcraft.com/products/pdf_files/multi.pdf
I have a 6 pin to handle the heaters with twisted 18awg wire and a 4 pin to handle 2 300 volt lines, audio ground, chassis ground. I used cable rated for the 300 volt and it's also has shielded that I have tied on one side to chassis ground.
I only did a separate power supply because of weight, The amp would be over a 100lbs if I had it all in one chassis. but the separate power supply made it a lot more expensive and a lot more work. I'm going to think twice before I ever do it again.
Last edited:
I did it for the same reasons as Davide. I built a tank of a PSU that was to be reused for a few different amps in the hopes of saving some money on the iron. The first was my dad's home-made 8417 PP amp which had a habit of eating power transformers. The second was an amp that never got off the bread board and I've since caught the SE bug. The PSU has since been somewhat dismantled for parts. Some day I'll get back to it.
Hi Guys
I do separate the psu and also separate the out-put trans from the out-put tube too, detail in my thread here( A new idea for power amp) people may not agree with me because of high voltage in cable, so far I built two of them one in 760v another in 800v with no problem atall.
Tony Ma
I do separate the psu and also separate the out-put trans from the out-put tube too, detail in my thread here( A new idea for power amp) people may not agree with me because of high voltage in cable, so far I built two of them one in 760v another in 800v with no problem atall.
Tony Ma
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.