• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Consensus on SMP's for indirect heaters (assumption is you want DC heaters)?

Jump to Latest
Windcrest77

Windcrest77

Member
Joined 2019
#1
I know this is an old subject, but... I was at Axpona last year and saw a tube phono pre-amp for sale in the $8,000 range, it used dual SMP's for the heaters! I've seen posters here who regularly use SMP's for indirect heaters. And for B+. Assuming someone does want to use DC... These off the shelf Meanwell supplies are just so dog-gone economical at ratings up to 29A! One would be tempted to just use them and save a lot of time.

No 6 volt units here but plenty of 12 and 24 volt where you could series connect like tubes as stereo pairs or quads. Using proper construction, distances, wiring, etc. Is there some absolute reason to not use a drop in an SMP module for an indirect heater? Or is it all opinion? Is the parasitic really that problematic?

line card:

MEAN WELL LRS Power Supplies Application and Features

Example supply:

LRS-100-12: MEAN WELL : <p>102W 12V 8.5A Single Output High Efficiency Power Supply</p> : Power Supplies & Wall Adapters
 
#3
Nothing wrong with them at all but switching supplies give you dirty looks around here. They are quiet as ever and only need some small caps and chokes for filtering. People are acting like the 8Khz switching supplies from the 1970s still exist.
 
#5
I have had excellent luck buying cheap AC-DC modules and adding an LC post-filter. I've gotten the switching spikes and ripple almost down to where I couldn't measure them anymore.

I haven't done it yet for a filament, but I did just buy a 20A 7.5V Meanwell switching supply for a DHT amp I'm designing and I'm going to see how clean I can make it. Nice thing on this one is there is an output adjustment. I don't have to worry about inductor DCR as I can compensate for losses by increasing the voltage a bit if needed.
 
#6
Not exactly SMPS but I'm using a switching regulator (LMR14030) for DC heaters (both indirect & direct heated). The switching frequency is 500kHz and the noise level is very low. The advantage over linear regulators is the low dropout so regular 6.3 VAC secondaries (with schottkys) can be used for 6.3 VDC.
 
#7
The Meanwell LRS units are quite nice. Always pay for shielding.

You can use the 7.5V Meanwell SMPS and just turn the adjustment pot all the way down, then you'll be just a bit above 6.3V. For fine tuning, you can add some series resistance to trim things down.
 
#8
I have been playing with a breadboarded 20+ WPC SE amp that runs two 36LW6 sweep tubes. Their heaters are powered with a LRS150-36.

I spent some time chasing down a low level high frequency "oscillation" that would randomly appear on the output of my amp in one channel only, and usually only when cranked pretty loud into speakers that were sitting on the workbench.

I never heard anything in the speakers, just kept capturing random HF bursts on the scope. Of course I suspected the SMPS, so I swapped it for an identical unit....now the "oscillation" was continuous, but still on only one channel.....The clip lead that I had used to ground the metal case on the SMPS was intermittent, swapping the supply made it totally open.

When using an SMPS for tube heaters, you need to oversize them. Different kinds need different amounts of oversize. The heater of a cold tube has a much lower resistance than it has when hot. The 450 mA current drawn by a 36LW6 is more than an amp cold. The LRS-150 will reliably start into 4 tubes, but is hit or miss into 6 tubes, and will not turn on at all with 8 tubes connected to it.
 
#9
@Windcrest77, I bet that you won't find a MeanWell SMPS inside the $8'000 preamp; maybe a XP power MTC or something else in the $250-$400 range, and possibly a custom design. There are several issues with the consumer grade generic SMPS you mentioned on post #1: they have a comparatively low MTBF, so they will be a reliability weak link in a tube amplifier design; they are not designed to power analog circuits, so they don't have a sufficient integrated filter on the input and output and are not shielded to reduce EMI. High efficiency is not a design priority for this kind of price-constrained modules, so expect a 15%-20% loss as heat that must be taken into account when placing the module under the chassis. You can use them on a tube amplifier as-is, and of course they will work. But I would only use a SMPS if there was a design advantage; cost is not the first goal in designing a tube amplifier today. There are standard SMPS modules that have about the same ripple and MTBF as a linear power supply with a significant reduction in weight an size, but generally have still a high price. It is coming down, but for now the best approach to optimize cost I know is to start with a shielded SMPS module (to keep the radiated EMI low) and add external filtering, calibrating it experimentally. This cannot be done without at least a good oscilloscope.
 
#12
Tubelab_com said:
When using an SMPS for tube heaters, you need to oversize them. Different kinds need different amounts of oversize. The heater of a cold tube has a much lower resistance than it has when hot. The 450 mA current drawn by a 36LW6 is more than an amp cold. The LRS-150 will reliably start into 4 tubes, but is hit or miss into 6 tubes, and will not turn on at all with 8 tubes connected
Click to expand...

I was thinking of possibly an over current circuit between the SM supply and filaments. Not a CCS supply, but over current protection where you set the ceiling at slightly above the most current needed after warm up maybe 500ma in your example. So current would never go above that while the tubes are warming up. Even if the SMPS is rated at 20A. Then after the tubes are warm the over current protection drops out leaving you with a pure voltage regulated supply, that just happens to be drawing whatever current is being demanded by a warm tube instead of a cold tube. Which is below the 500ma limit you dialed in.

Along with over current any additional filters you want in the chain too.

I don't know, you guys are light years ahead of a kit builder like me who is just trying to learn a little first-principle design here and there.
 
#13
Most switch mode supplies have over current protection built in. If you had a 36V/20W SMPS and you tried to heat a 36LW6 with that, you'd never get there. The supply would just pulse into the tube heater endlessly (been there, done that!).
 
#14
The MTBF number must not be interpreted as literal value; it is inflated like the PMPO peak musical power output number. Board life cannot be greater than the life of the capacitors, for instance. MTBF may be useful as a reference. The MTBF of the higher quality SMPS I quoted on my previous post is 1'700'000 hours. There is a known industrial machine manufacturer that a while ago had the idea to fit this MeanWell power supplies range on its control panels. I can ensure you that they don't survive 73 years.
 
You must log in or register to reply here.
Top Bottom