Hey you all, just looking for perspective here. I'm collecting parts to monoblock SE 2A3's and understand full well that 83 rectification (800V/250mA max) is gross overkill. That said, I have a bunch in my stash, but wonder about what if any negative sonic impact might result. As info, years ago I had a pair of Loesch Triple (parallel) 2A3's outfitted with 83's that sounded, well, pretty damn good as you might expect from the designer. Appreciate any advice you might offer.
That depends entirely on how you build it. The traditional issues with MV rectifiers need to be addressed. That usually means a shielded rectifier and appropriate RF decoupling. Do that and there shouldn't be any negative impacts. Other than having to wait for the rectifier to warm up before you can listen to your amp. 😉
Thanks for the reminder in re shielding and otherwise dealing with RF issues. Hadn't gone down that path in a long time and resisting the temptation to do a 5AR4. Thumbs up. 👍
Remember that mercury vapor rectifiers emit copious amounts of UV radiation and need about 30min to heat properly.
Not sure if wearing sunglasses and sunscreen during use has much sonic effect, but it may...
Not sure if wearing sunglasses and sunscreen during use has much sonic effect, but it may...
Thanks! I should be safe. I always wear sunglasses while listening. 😎 I find that they improve bass performance by perhaps 20 - 25%. 30% on Saturdays. That said, thanks for the reminder about the pre-heating time. 15 to 30 minutes, as I understand it, before applying B+. Inconvenient, true. But I can always go back to a more conventional rectifier if it becomes too bothersome. Thanks for the response.
Mercury vapor rectifiers do indeed produce UV radiation.
Regular glass however, is opaque for the most harmful short wavelength UVB and UVC.
Glass is semi-transparent for UVA ('blacklight') so some of that will get through.
Would be nice to know if it is enough to light up white clothing like in nightclubs 😎.
Regular glass however, is opaque for the most harmful short wavelength UVB and UVC.
Glass is semi-transparent for UVA ('blacklight') so some of that will get through.
Would be nice to know if it is enough to light up white clothing like in nightclubs 😎.
Mercury Vapor Rectifiers are pretty.
Mercury Vapor Rectifiers can be pretty dangerous if they break while they are hot.
They have low voltage drop through the tube.
Solid state rectifiers are not pretty.
Solid state rectifiers have lower voltage drop than mercury vapor tubes.
A resistor can increase the voltage drop to be like a mercury vapor tube.
Do whatever you want, you can compensate, very easy if you do it before you purchase the power transformer.
The above are all my opinions, and may not be worth a plug nickel.
I use quiet solid state rectifiers, and choke input power supplies, by careful selection of all the parts.
I rely on the output tube for the pretty factor.
Mercury Vapor Rectifiers can be pretty dangerous if they break while they are hot.
They have low voltage drop through the tube.
Solid state rectifiers are not pretty.
Solid state rectifiers have lower voltage drop than mercury vapor tubes.
A resistor can increase the voltage drop to be like a mercury vapor tube.
Do whatever you want, you can compensate, very easy if you do it before you purchase the power transformer.
The above are all my opinions, and may not be worth a plug nickel.
I use quiet solid state rectifiers, and choke input power supplies, by careful selection of all the parts.
I rely on the output tube for the pretty factor.
They are very dangerous. Only one killed my " Spender " woofer . Therefore, I'm willing to buy two of your many to kill the other one🙂 Sometimes to move on from the speaker it takes death to do us apart🙂 I'm serious..
I'm currently building a headphone amplifier with an 83. The powersupply has a 30 second delay, bypass switch for an infinite delay should I have swapped the rectifier or moved the amplifier, RF chokes on the 83, a choke, an E-choke and some ferrite beads here and there. Not gotten to finalizing it yet but you gotta suffer a bit for those looks 🙂
I seem to recall a previous thread discussing the warm up for the 83; think there was reference to a valve tester that used the 83 and had no delay. For most of the bigger MV the first use after storage requires a much longer warm-up of up to 30 minutes as the mercury gets everywhere, then the daily in service warm up is <30 seconds. The small 83 may be fine without as I don't see it mentioned in the datasheet. Newer production tended to have less mercury dumped inside so probably have less risk of arcing in those.
For me the niggling issue with MV are that I understand the gas is odourless and unlike vacuum tubes there is no getter to turn white, so it's not easy to spot leaks; the last thing I want is to get slowly gas poisoned in my home. I first test them outside the living space, and check for cracks in the glass around to lead-out. I also check the colour of the filament and current draw. If they don't reach dull red filament or draw too much current then I conclude that gas has got in there and reject them.
I gather the RF hash conducted and emitted applies more to noisy environments like transmitters where they were originally used. I didn't think it was a big issue in audio. Perhaps best to bread board and play with transformer orientation and rf chokes.
I don't think 83 is overkill at least no more than the 5AR4 you mention. MV prefer to be run Choke Input, although I see Cap input mentioned in the RCA datasheet for the 83. I wouldn't use 83 to obtain 800v(dc filtered?), but it's fine for 300vdc in full wave choke input as PIV will be around 1Kv, the rated PIV being 1.5Kv. If you don't have a suitable CT transformer a bridge adding two single diodes, the slow warm-up of damper diodes may provide all the delay you need. I think I may have seen an example like that on Thomas Mayer's blog, Vinyl Savor.
For me the niggling issue with MV are that I understand the gas is odourless and unlike vacuum tubes there is no getter to turn white, so it's not easy to spot leaks; the last thing I want is to get slowly gas poisoned in my home. I first test them outside the living space, and check for cracks in the glass around to lead-out. I also check the colour of the filament and current draw. If they don't reach dull red filament or draw too much current then I conclude that gas has got in there and reject them.
I gather the RF hash conducted and emitted applies more to noisy environments like transmitters where they were originally used. I didn't think it was a big issue in audio. Perhaps best to bread board and play with transformer orientation and rf chokes.
I don't think 83 is overkill at least no more than the 5AR4 you mention. MV prefer to be run Choke Input, although I see Cap input mentioned in the RCA datasheet for the 83. I wouldn't use 83 to obtain 800v(dc filtered?), but it's fine for 300vdc in full wave choke input as PIV will be around 1Kv, the rated PIV being 1.5Kv. If you don't have a suitable CT transformer a bridge adding two single diodes, the slow warm-up of damper diodes may provide all the delay you need. I think I may have seen an example like that on Thomas Mayer's blog, Vinyl Savor.
Using two damper diodes in a hybrid mercury vapor (MV) and vacuum diode bridge can indeed provide the necessary warm-up delay. Thomas Mayer showcased this approach in some of his earlier blog entries, definitely worth checking out.
Regarding the use of an 83 tube in valve testers without a delay circuit: it's worth remembering how those testers were typically operated. After powering on the tester, users would often spend a bit of time inserting and configuring the tester before any meaningful current was drawn. That delay, even if unintentional, gave the 83 ample time to warm up properly. So it’s a good idea to take that context into account before directly translating its use into an amplifier circuit (which applies a load more or less straight away).
I have the RF chokes doing nothing in a drawer so with the approach of "what harm could it cause" I bolted them to my chassis.
You do pose a good point about it not being immediatly visible if such a MV rectifier developed a leak but I'd suppose the heater would burn out in no time if air is introduced?
Regarding the use of an 83 tube in valve testers without a delay circuit: it's worth remembering how those testers were typically operated. After powering on the tester, users would often spend a bit of time inserting and configuring the tester before any meaningful current was drawn. That delay, even if unintentional, gave the 83 ample time to warm up properly. So it’s a good idea to take that context into account before directly translating its use into an amplifier circuit (which applies a load more or less straight away).
I have the RF chokes doing nothing in a drawer so with the approach of "what harm could it cause" I bolted them to my chassis.
You do pose a good point about it not being immediatly visible if such a MV rectifier developed a leak but I'd suppose the heater would burn out in no time if air is introduced?
I've had a few that on first test were too dull and didn't get so bright, but they've had decades to slowly leak air in or could have come off the production line like that. What I don't know is how high the pressure gets inside during operation nor how capable mercury vapor is at squeezing through the gas/metal seal. I seem to recall service life being in the thousands of hours so that's somewhat reasurring. My baseless suspicion for a tube seeing intermittent service is that where there is a weakness MV will get out before air gets in. The more I've learned about them, the less worried I am about using them; a few years ago I would have refused to consider using them. The organic mercury in fish per unit is far more toxic than the element., and I eat plenty.