1'st off I can read tube data and everything I read says not to exceed 4uf after a 5R4GY tube rectifier. How does voltage and mA draw factor into that? My amp has 85mA draw and operates below 300 volts H.V. supply. I also am using a Thermistor to allow a slower start up.
I see no arcing with a 20uF poly cap in the 1'st position of the CLC filter.
I made the mistake listening to a 5R4GY vs a 5u4G and I love the sound of this 5R4GY. I could drop the cap down in size if needed at the expense of lowering voltage.
Walt
I see no arcing with a 20uF poly cap in the 1'st position of the CLC filter.
I made the mistake listening to a 5R4GY vs a 5u4G and I love the sound of this 5R4GY. I could drop the cap down in size if needed at the expense of lowering voltage.
Walt
The amp is a Tubelab SE with 45 output tube.
The very good sounding ar4's are a little above my pay grade. the RCA 5E4GY mil spec tube can be had for $30.00 to $50.00 and sound better to me the the RCA 5U4G's I have using.
The very good sounding ar4's are a little above my pay grade. the RCA 5E4GY mil spec tube can be had for $30.00 to $50.00 and sound better to me the the RCA 5U4G's I have using.
The problem is peak rectified current during warm up when the cap is charging. Possibly your NTC is limiting current enough to limit the inrush current to a value the rectifier can survive.
The 5R4 has a lot more forward drop than a 5AR4 and significantly more than the 5U4. You might want to measure the supply voltage because you have significantly reduced output voltage..
I would recommend reducing that capacitor value to about 5uF, even 10uF will improve rectifier life.
The 5R4 has a lot more forward drop than a 5AR4 and significantly more than the 5U4. You might want to measure the supply voltage because you have significantly reduced output voltage..
I would recommend reducing that capacitor value to about 5uF, even 10uF will improve rectifier life.
The 4uF capacitor value is when the rectifier is running at maximum values; the fact that you are less than 1/2 the MaxPV and about 1/3 the MaxPmA means that you can allow the rectifier to see a higher capacitance before the choke. You could probably get away with a 10uF cap without arcing, but you might be gambling slightly with 20uF.
Personally, if I'm running that low of mA and can handle the additional B+, I'd rather use a 5V4G, but it sounds like you are attached to the 5R4GY.
-D
Personally, if I'm running that low of mA and can handle the additional B+, I'd rather use a 5V4G, but it sounds like you are attached to the 5R4GY.
-D
That 4uF is for typical operations in 1948. The old 5U4G data said the same thing but by 1956 the typical operation was speced for 40uF, and they were the rectifier for Maggie amps with a 40uF input cap. I've not seen a 5R4 data sheet specify a max do not exceed cap rating. It's all about high AC voltage, the load and internal resistance of the PS. I've used them in my ST-70 with 30uF @ 200ma. and in my Maggie with 40uF @ 130ma. No problem, and they pull the B+ down a bit to compensate for high line voltage.
Thanks for the feedback. I will say from a clod start it takes about 16 seconds for the voltage to build enough for the 5R4gy to start making music. I looked close at the tube data for and at the end it stated that if a larger cap was selected it was recommended to increase the plate impedance. Can I assume they are indicating adding resistance? And if that is correct can I pad the center tap with a resistor?
I am not stuck on the 5r4gy, and in this case the voltage is still in the good range (270 volts).
I am open to alternatives beside the 5u4G.
I am not stuck on the 5r4gy, and in this case the voltage is still in the good range (270 volts).
I am open to alternatives beside the 5u4G.
The additional resistance, if you believe you'll exceed the max current rating, should go in each of the AC leads ahead of the rectifier . I wouldn't put it at the CT. That is like lifting the CT and having a single winding secondary.
There was recently a similar discussion about this on another forum which can shed some light on the first cap issue.
https://www.audioasylum.com/forums/tubediy/messages/25/254231.html
https://www.audioasylum.com/forums/tubediy/messages/25/254231.html
I was actually looking to buy a Mullard CV593 and did not want to risk destroying an expensive tube but it looks to be good.
In an amplifier I did use the 5R4GY with two 47μF caps series connected, but with a time delay, no problem at all.
Once, the only cap on hand was 220μF, and the valve did some little arcing at startup, but no problem, although I do not recommend.
As a general rule, you can somewhat exceed one of the rectifier limits provided you stay well inside all the other limits. If your current, voltage and series resistance are not on the edge of what is allowed, then you can have a slightly bigger reservoir capacitor. Softstart gives a bit more room.
When adding series resistance it is traditional to add it to each anode, not the CT. I believe the reason for this is to limit current flow if an arc occurs; a CT resistor will not necessarily do this.
When adding series resistance it is traditional to add it to each anode, not the CT. I believe the reason for this is to limit current flow if an arc occurs; a CT resistor will not necessarily do this.
The diode datasheet values can be used to set up a PSUD2 simulation, using the default 5R4GY diode model. For the design max levels of 900VAC and 575 ohm and 4uF and 150mA load, my sim indicates a load of 6.2 kohm, 925VDC with a 1A pk hot start and 0.5A peak continuous.
You can then play around with different power transformer voltage and effective resistance, and first filter capacitance, to indicate if you are stressing the diode more so than its default max design levels.
There is some risk with a NTC - it may well limit startup peak current, but that may assume a cold NTC. Somewhat similar to valves remaining hot, an NTC will also remain hot, so a 'hot start' scenario may not be easy to always avoid.
You can then play around with different power transformer voltage and effective resistance, and first filter capacitance, to indicate if you are stressing the diode more so than its default max design levels.
There is some risk with a NTC - it may well limit startup peak current, but that may assume a cold NTC. Somewhat similar to valves remaining hot, an NTC will also remain hot, so a 'hot start' scenario may not be easy to always avoid.
Last edited:
Old U.S. made rectifier tubes are exceptionally high quality and are extremely difficult to kill, even if you try. I don't think I've ever seen a bad 5R4. I have, literally, a tub full of good rectifier pull outs, 5R4, 5U4, 5Y3, 5V4, etc., that outlived the equipment they were installed in.
The 47uF cap in the SSE or TSE will be fine. Don't worry about it. If you have the one in a million tube that can't handle it, just get another.
Win W5JAG
- Status
- Not open for further replies.