|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.6.253.18
This is a follow-up of a thread started by Rick R on 7-5-08. (see seven threads below)It seems only natural to me that a rectifier tube should have balanced plates for the cleanest output waveform with lowest noise and ripple.
One thing I have never understood, though: Why are the English settings on a Hickok tester different for the two plates?
With a 5U4, for example, the English setting for plate #2 (the 0400 setting), is 30, whereas on plate # 1 (the 0600 setting),the English setting is 35. This seems to hold true on all filamentary rectifiers, including 5U4, 5AS4, 5V3, 5Y3, etc. All rectifiers with a cathode (5AR4/GZ34, 5V4, etc.) have the same English setting for both plates.
The one exception seems to be the 5R4, which has no cathode, yet both plates have the same English setting.
I asked this question years ago on this forum. Even Hickok guru Chris Haedt (may he rest in peace)did not know the answer.
Can anyone share some insight?
Thanks in advance! Al
Edits: 07/06/08Follow Ups:
Seems obvious to me... there's two heaters in series, 2.5V between the center points. So there's a greater voltage across half of the tube. Reverse the heater connections and you'll see the OTHER half reads higher. The tube tester uses a relatively low voltage, so this effect is much greater than in a typical circuit. If the B+ is taken from a center-tap on the 5V winding, this effect disappears (along with a small amount of 60 Hz ripple).
The typical Hickok rectifier test circuit uses a tapped 35VAC winding to provide both filament and plate voltage. Therefore, on average, one plate-cathode element will "see" roughly 34VAC while the other will see ~31 VAC. This is a result of the voltage drop across the filament. Hickok usually accounted for this by either giving a different "shunt" or multiplier value for each plate in an attempt to provide equal noninal or reject values on the meter. An indirectly heated cathode type rectifier such as a 5V4 doesn't have the filament voltage drop issue and Hickok's setup reflects this.
Every std Hickok tester I've seen gives a different shunt/multiplier value for the 5R4 so I suspect that Alpha Al's set-up info is in error. Hickok's set-up info was known to be in error once in a while.
IMO, this discussion brings up a really major point about testing and comparing power rectifiers with a std Hickok tester circuit: With less than 35VAC on the plate, take all readings with a huge grain of salt and don't expect to find arc prone tubes with it.
Hi Tom: Makes sense. But why should 5R4GY have the same English settings on both plates? It is essentially same as 5U4, but higher max voltage. Hickok's error?
I believe the tester designers knew that rectifiers would have varying values of results, as their testing does not quite simulate a tough load on the tubes. Furthermore, these minimum acceptable values of tube results are nominal, not absolute. We all know that weak testing tubes can still last a good test of time,(and sonics), in our audio gear. As long as the tube tests decently, and closely related for the two results, I would not dwell on it, just enjoy....
I wonder if anyone has tested just how symmetric a solid state bridge rectifier or set of diodes really is? My guess is that under load the components may show asymmetry, also.
Just a WAG.
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: