Hi all,
I could use some advice.
I plan to build a clone of a Fender 5E5 Pro amplifier using, among other things, the power and output transformers from a Bell Carillon 35A PA amplifier. Both amps use push pull 6L6s.
The difference, and the reasonI need help is, the Fender used a 5U4 rectifier; the Bell used a solid state full wave bridge and does not have a 5V winding.
So I'll be forced to use a solid state bridge on the Fender clone I'm making.
My question: What can/should I do to limit inrush current with this kind of rectifier? I am sure the Bell had an allowance for it (I could never find the schematic) but I know the Fender, designed for a tube, might be problematic if I don't limit the incoming current one way or another.
Any suggestions? Thanks!
Ted
I could use some advice.
I plan to build a clone of a Fender 5E5 Pro amplifier using, among other things, the power and output transformers from a Bell Carillon 35A PA amplifier. Both amps use push pull 6L6s.
The difference, and the reasonI need help is, the Fender used a 5U4 rectifier; the Bell used a solid state full wave bridge and does not have a 5V winding.
So I'll be forced to use a solid state bridge on the Fender clone I'm making.
My question: What can/should I do to limit inrush current with this kind of rectifier? I am sure the Bell had an allowance for it (I could never find the schematic) but I know the Fender, designed for a tube, might be problematic if I don't limit the incoming current one way or another.
Any suggestions? Thanks!
Ted
1st, this belongs on the Instruments and Amps "board".
The amount of delay provided by a directly heated rectifier, like a 5U4, is quite small. Only types with cathode sleeves, like the 5AR4, really slow B+ rise.
A CL-130 inrush current limiting thermistor between the SS bridge rectifier and the PSU filter will give you about the same delay as a 5U4 provides. You will have to "play" with resistors to get the "sag" associated with vacuum rectification.
The amount of delay provided by a directly heated rectifier, like a 5U4, is quite small. Only types with cathode sleeves, like the 5AR4, really slow B+ rise.
A CL-130 inrush current limiting thermistor between the SS bridge rectifier and the PSU filter will give you about the same delay as a 5U4 provides. You will have to "play" with resistors to get the "sag" associated with vacuum rectification.
Thanks Eli. And apologies on putting this on the wrong board. I appreciate the help.
It is probably limited by the transformer anyway.
I have designed an 80 watt valve amp and used SS rectification and it worked fine with out a thermistor etc.
I have designed an 80 watt valve amp and used SS rectification and it worked fine with out a thermistor etc.
Hi. I didn't worked with thermistors yet but as I understand have to choose it in respect with amount of current the circuit draw to achieve its minimum resistance when heat it if you want to keep it permanent in circuit. A thermistor designed to work at 10A will not present the same resistance when heat if you draw 1A through it. So I don't know if bigger is better cause you get more residual resistance series with the circuit. Just think the good aproach is to choose one rated for the nominal current of your circuit but look for the peak current specification at start.up. Also look for working voltage. Thermistors rated for high voltage range are crazy expensive.
Last edited:
You could also use a series current limiting power resistor which would be bypassed with a relay or transistor after the inrush was over, i.e. when the power supply voltage had risen sufficiently. Needs a few extra components to drive the relay/ transistor switch, but works rather well.
The only things that normally suffer from inrush are the rectifier and the fuse. SS rectifiers can handle gobs of inrush, so no trouble there, and premature fuse failure is a minor inconvenience. What's more, you are copying a design that originally had a tube recto, i.e. small filters caps, so the inrush should be a lot LESS than what it would normally be with an SS recto design. Use a thermistor if you like, but don't spend time worrying about it.
The filter caps might be smaller, but the voltage is higher too. Tube rectifiers also tend to have higher ON resistance. Also, the warm up time for a tube rectifier might be small for a tube, but it is a LOT more than that of a solid state diode. 😉
I have had solid state rectifiers die on me when dropped in to replace a blown tube, so the OP's fears are fully justified. If the tube data tells you the effective series resistance at rated current, then a resistor of similar value in series with the SS rectifier should be the minimum protection used.
thanks Merlin. I've pretty much let it be.
I was worried about that inrush as the voltage goes immediately to 700+ at the first filter cap before the tubes heat up, and drops off to about 440. Though I've got a 500V cap in that spot I was worried that that initial rush would fry it. Everything seems to be operating OK though (so far) 🙂
I was worried about that inrush as the voltage goes immediately to 700+ at the first filter cap before the tubes heat up, and drops off to about 440. Though I've got a 500V cap in that spot I was worried that that initial rush would fry it. Everything seems to be operating OK though (so far) 🙂
That will be because of the reverse voltage stress, not because of inrush current.
That is not normal...
Last edited:
While there are no definitive studies I've seen of "cathode stripping" of signal and power tubes due to instant on SS rectifiers, it is quite apparent that B+ ramps up a lot faster with SS than with a tube rectifier.
I've been sticking with tube rectifiers since many of my tubes are no longer manufactured, like 7199.
However, for $3 you can simulate the gradual turn on of a tube, with a modern component. GE CL-xxx NTCR's (negative temperature coefficient resistors) are about $2, and the cinch solder terminal strip to mount them on is about $1. I put them in series with the mains winding of the power transformer after the fuse and power switch. Datasheet is on datasheetcatalog.com, for example the CL-90 I used for my preamp. You size the part depending on the full power current you want, which is the hot value. Newark.com stocks them in the US; known as farnell overseas.
Relays timers and series resistors shorted out by the relay are incredibly clunky for this kind of need when one part will do it. The e-cap in the timing circuit us usually a amp killer in the 20 year timeframe. Many $$$$$$ electric organs are killed at 20-30 years age by the e-cap in this soft turnon/silence circuit.
I've been sticking with tube rectifiers since many of my tubes are no longer manufactured, like 7199.
However, for $3 you can simulate the gradual turn on of a tube, with a modern component. GE CL-xxx NTCR's (negative temperature coefficient resistors) are about $2, and the cinch solder terminal strip to mount them on is about $1. I put them in series with the mains winding of the power transformer after the fuse and power switch. Datasheet is on datasheetcatalog.com, for example the CL-90 I used for my preamp. You size the part depending on the full power current you want, which is the hot value. Newark.com stocks them in the US; known as farnell overseas.
Relays timers and series resistors shorted out by the relay are incredibly clunky for this kind of need when one part will do it. The e-cap in the timing circuit us usually a amp killer in the 20 year timeframe. Many $$$$$$ electric organs are killed at 20-30 years age by the e-cap in this soft turnon/silence circuit.
Last edited:
Before the amplifying tubes warm up, they are not drawing any current from the B+ rail, so the initial B+ voltage will be higher than expected. My only experience with tubes is with X-ray equipment, and in those, such a condition causes arcing as the maximum voltage rating of the tube is exceeded. That probably doesn't apply here.
I get indianajo's point about complexity, although it is usually done by voltage (about 80% of full voltage) rather than with a timer. Most SMPS use this technique. NTCs work well, but are not very efficient (which is why they are seldom used in an SMPS), but tube guys don't much care about that.
One thing to keep in mind is that NTCs get hot; in fact they work because they get hot. Treat them like a power resistor - stand off; air circulation, not touching anything that can melt or burn, etc.. Allow sufficient time for them to cool after turning off the amp before turning it back on, or the still hot NTC will not do it's job of limiting the inrush.
I get indianajo's point about complexity, although it is usually done by voltage (about 80% of full voltage) rather than with a timer. Most SMPS use this technique. NTCs work well, but are not very efficient (which is why they are seldom used in an SMPS), but tube guys don't much care about that.
One thing to keep in mind is that NTCs get hot; in fact they work because they get hot. Treat them like a power resistor - stand off; air circulation, not touching anything that can melt or burn, etc.. Allow sufficient time for them to cool after turning off the amp before turning it back on, or the still hot NTC will not do it's job of limiting the inrush.
Peavey wrote an excellent article on why you should use an inrush current limiter on most tube amps, solid state or tube rectification.
This article starts out explaining the "standby" switch on tube guitar amplifiers, and what it is really for. (Hint, no-one here mentioned it, but it will be interesting to read for this discussion.)
This article starts out explaining the "standby" switch on tube guitar amplifiers, and what it is really for. (Hint, no-one here mentioned it, but it will be interesting to read for this discussion.)
That article has been debunked before. It's just received wisdom drivel.
Last edited:
Use the search function (try the Tubes/Valves forum). Standby switches, cathode poisoning etc have all been discussed to death. Here's a beginner's precis:
The Valve Wizard
The Valve Wizard
- Status
- Not open for further replies.