Max power test - JJ GZ34 vs Eico Gt Britain (Mullard?) GZ34 vs Solid State Rectifier Distortion performance in the same Dynaco ST-70

knotscott said:
The stories of modern rectifier tubes going up in smoke are exactly why I opted for SS rectification in my Dyna/VTA 70s. I just didn't see any benefit to offset the risk. Roy Mottram's SS rectifiers have two diodes and a 12102 thermistor.
Click to expand...
I left the yellow sheet mod in place when I went with a SS rectifier in my VTA
ST-70. I chose the Copper top due to it having resistors to mimic the 17v drop of a 5AR4. I used the Copper tops that mimic two 5AR4's in my M-125 mono blocks. Both Copper tops I used also have thermistors inside as well. So far they've held up just fine in my mono blocks since building thwm several years ago. I see no reason to use a tube rectifier in these amps other than if you just want some added tube glow over performance.
I'm not worried about the faster ramp up of the B+. It hasn't been an issue with my ST-35, no thermistor in it, or the other amps amp preqmps I've built.
 
Lavane said:
I left the yellow sheet mod in place when I went with a SS rectifier in my VTA
ST-70. I chose the Copper top due to it having resistors to mimic the 17v drop of a 5AR4. I used the Copper tops that mimic two 5AR4's in my M-125 mono blocks. Both Copper tops I used also have thermistors inside as well. So far they've held up just fine in my mono blocks since building thwm several years ago. I see no reason to use a tube rectifier in these amps other than if you just want some added tube glow over performance.
I'm not worried about the faster ramp up of the B+. It hasn't been an issue with my ST-35, no thermistor in it, or the other amps amp preqmps I've built.
Click to expand...

Same. I left the yellow sheet mods in place as well....way too much work to take them out!
 
mcgjohn said:
Tube rectification removes more of the switching noise than SS rectifiers do.
Click to expand...
What you are talking about is a swept resonance, caused by the capacitive aspect of the rectifier junction interacting with the inductance of the transformer. There is an electrical peak that can form which can go into oscillation; thus the 'switching noise'.

A proper snub, consisting of a single resistor in series with a small capacitance, placed across the input to the rectifier, can shut that down. Bypassing the rectifier has almost no effect. Once you have the rectifier neutralized in this manner it can be very quiet.

When snubbing the noise its helpful to keep the power transformer leads as short as possible, since what you are really doing is snubbing the transformer rather than the diodes.

You could also use HEXFRED rectifiers, which are ultra fast recovery and ultra soft recovery. When I use those its rare I have to snub anything.

The elephant in the room is that tube rectifiers have a filament. When the AC mains Voltage goes down, the rectifier cools off and so has a higher Voltage drop across it. Conversely when the mains Voltage is higher, the rectifier has a smaller Voltage drop, in addition to the increased Voltage coming from the power transformer high Voltage winding.

Its like trying to balanced on top of a ball; a double whammy, making the amp a lot more sensitive to line Voltage. IMO tube rectifiers do not have an application in home audio but are nice in guitar amps if you play the blues. This is one application where tubes are simply outclassed by solid state.
 
Lavane said:
I left the yellow sheet mod in place when I went with a SS rectifier in my VTA
ST-70. I chose the Copper top due to it having resistors to mimic the 17v drop of a 5AR4. I used the Copper tops that mimic two 5AR4's in my M-125 mono blocks. Both Copper tops I used also have thermistors inside as well. So far they've held up just fine in my mono blocks since building thwm several years ago. I see no reason to use a tube rectifier in these amps other than if you just want some added tube glow over performance.
I'm not worried about the faster ramp up of the B+. It hasn't been an issue with my ST-35, no thermistor in it, or the other amps amp preqmps I've built.
Click to expand...
WS1?

WS1 Copper Cap – Weber Speakers

www.tedweber.com www.tedweber.com
 
Mr Boochie said:
I used the Weber Copper Top W68. It replaces amps with dual original GZ34s.

The WZ34 is to replace amps with a single GZ34.
Click to expand...
The ST-70 was mentioned is why I asked.

Their site says not to use ""FOR USE IN GUITAR AMPS. Not designed for use with HIFI tube amps, tube radios, etc. – Try the WS1 instead.Drop-in Replacement for GZ34/5AR4. Voltage drop: 17V at Maximum Current: 225ma.""

Just asking..
 
atmasphere said:
It would be nice if someone had a solid state rectifier that had a fixed Voltage drop regardless of current draw- IOW, regulated...
Click to expand...
The Weber does have a fixed voltage drop resistor. So you want a variable voltage drop relative to the current demand to simulate the tube?

Solid-state diodes will exhibit a voltage-drop spectrum (albeit not as pronounced as tube rectifiers) with current demand.

The only way to simulate the complex nature of a specific tube's voltage sag with current demand is to use a tube. It's why they are still popular in small to moderate-sized guitar amps. :)
 
a resistor's voltage drop depends purely on current loading. I think he's meaning something that drops a flat 30 (or whatever) volts regardless of input. It probably wouldn't be that difficult to design the circuit but offhand I don't know how to pull it off without a connection to ground. I wouldn't trust just sticking a zener in series with the rectifier diodes since it would see a whole lot of ripple and they aren't really designed for that sort of duty.

With a connection to ground, I'd say probably rectifiers to the source of a mosfet. Zener from source to gate, with a resistor to ground. That should put the gate however many volts below rectifier output voltage. Drain out to the rest of the circuit. Still probably sub-optimal with all of the ripple. That would fit a lot better after the first filter cap, but thats definitely not going to cut it as a plug-in. Also not entirely sure that a fixed voltage drop would really be ideal. If you're going to the bother of regulation, wouldn't an actually regulated voltage be the goal, not just whatever comes out of the rectifier minus some fixed number?
 
Okay. No dropping resistor in the WZ1. That would be a straight 1N4007 sub (or some equivalent).

I have had good luck with the WZ34. But watch the B+. It may still increase, even with the built-in resistor.
 
Mr Boochie said:
Okay. No dropping resistor in the WZ1. That would be a straight 1N4007 sub (or some equivalent).

I have had good luck with the WZ34. But watch the B+. It may still increase, even with the built-in resistor.
Click to expand...
I don't remember what I got for voltages, but I did try just a couple diodes in a tube base, and the WZ-34 and checked the voltages between the two when I bought the WZ-34.
I do remember the B+ being lower than the two diodes though so I left the WZ-34 in the amp. All my lytic's are 500V.
 
atmasphere said:
A proper snub, consisting of a single resistor in series with a small capacitance, placed across the input to the rectifier, can shut that down. Bypassing the rectifier has almost no effect. Once you have the rectifier neutralized in this manner it can be very quiet.
Click to expand...
Imho, there is no need to add two half-winding snubbers when using a valve rectifier, as the turn-on and turn-off transitions are so soft that any hint of commutation related noise is pretty much unmeasureable. Even with only plain 1N4007 rectification, the commutation noise is likely way down in the weeds, which is helped by the high voltage winding DCR and its self capacitance. I was barely able to discern the difference against the low noise floor using a soundcard REW spectrum. When identified, poor noise performance is typically related to poor layout and wiring.
atmasphere said:
The elephant in the room is that tube rectifiers have a filament. When the AC mains Voltage goes down, the rectifier cools off and so has a higher Voltage drop across it. Conversely when the mains Voltage is higher, the rectifier has a smaller Voltage drop, in addition to the increased Voltage coming from the power transformer high Voltage winding.
Click to expand...
Have you measured the change in a B+ level for say a 5% drop in just the 5V heater voltage of a bogey amp? Without some extra detail of conditions, I think this type of comment has no tangible merit to differentiate between valve and ss rectification.
 
Lavane said:
I don't remember what I got for voltages, but I did try just a couple diodes in a tube base, and the WZ-34 and checked the voltages between the two when I bought the WZ-34.
I do remember the B+ being lower than the two diodes though so I left the WZ-34 in the amp. All my lytic's are 500V.
Click to expand...
I know of several people who went to the WZ34 with ST-70 (or similar-power, vintage amps). No problems that I heard of. For sure, they like not having to find vintage NOS Mullard or US-made GZ34/5AR4 tubes anymore.

The WZ34 is a godsend in the original Vox AC-30s. Those amps eat the Mullards. :(
 
Rectifier voltage versus current characteristic likely shows more compression of the 'top' end (highest current passed during each mains cycle) as a rectifier tube ages and cathode emission wanes. The typical test for cathode emission (life) is to measure the change in a parameter for nominal and -5% (or -10%) heater voltage - certainly a typical test for output stage valves to see if they are approaching end-of-life. It's likely that there is negligible parameter change between nominal and +5% heater voltage, which would tend to make any comparison dependent on the amp and what normal voltage is on the 5V heater, and how much life was left in the valve.

Rectifier diodes, by design, also have to accommodate long term operation at their rated continuous peak current level, even though they have a very much higher short-duration peak current level capability - which sort of indicates how much spare cathode emission capability is designed in. If different rectifier models are compared, then as brought up in comments in facebook and this thread, there are a few aspects that are likely mixed together and may get misinterpreted in the rush to come out with a simple interpretation of measurements. Perhaps the main aspect highlighted is the B+ level, and the tested output power (and how they could be interacting and misconstruing the results). Another aspect not really mentioned is the THD difference between channels. Also, the presented waveforms may be a little misleading, as they are not knowingly taken at the same time during a mains cycle (eg. at the portion of time when rectifier current is at or near its peak), and one set of waveforms suggest the rectifier valve may have significant imbalance (which may just be a single poor sample, or alternatively representative of the model, but no way of knowing without more sample testing).
 
Last edited:
gadget73 said:
or parallel tube rectifiers at least, but that adds expense. At some point a single rectifier just isn't' up to the job though


Roy Mottram is offering a power supply board for these now that uses a pair of diodes. I bought one for a project, though I'm not going to use the single 1N4007 called for. Those just don't have the PIV needed. I picked something else, drawing a blank right now on the part number, but 1.6 KV PIV for like 17 cents each.
Click to expand...
the erhard ultimate 7o or ultimate upgrade boards, now sold by miller audio, include ss rectificied PS.
 
You must log in or register to reply here.
Back
Top Bottom