My take on a heater supply...

What people are working on at the moment
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#1 My take on a heater supply...

Post by jack »

Well...

This is something I've been working on-and-off on for about a year - I mentioned it in passing to Nick (ISTR) in a pm about other stuff, but it was but a glimmer in my eye at that point.

An ultra-low-noise, ultra-high efficiency, extremely linear slow-start heater supply for any voltage from 5V up to 30V and up to 12A. Fully adjustable ramp time, true precision linear current startup, precision current monitoring available too... however it's almost all SMD as most devices are not available in PTH packaging. The prototypes I built sound very very clean to me, so it's not just about the engineering, it's about the sound.

What it comes down to is that I really don't like R/C slow-start DHT/heater supplies much - the current ramp is not very linear, can be front-loaded, often doesn't cater for the non-linearity of the filament and the bridge rectifier is an arcane beast that in this configuration (low voltage, high current) burns energy.... plus a bunch of other stuff.

Clean sheet of paper time. Time to think outside of the box a bit...

It just so happens that a few years ago I became aware of Ideal diode bridges, an article in EDN Design Ideas "Linear ramp generator uses one op-amp" (26/01/15), another from 21/08/13 called "Multiple PSUs share load" and a bunch of other stuff. I've attached both EDN articles to this post. I played around with discrete ideal bridges for my Tesla Coil projects, and then thought about audio...

A really good current sense amplifier for this is the INA250, an astonishing device with inbuilt precision shunt for either high or low side sensing and good for up to 15A.

All these added up to an interesting project. I strongly recommend folk read up about ideal diode bridges - see the LT page at http://www.linear.com/products/ideal_diode_bridge and specifically the data sheet on the LT4320 at http://www.linear.com/product/LT4320

I've written a Excel sheet to calculate the component values for the ramp generator and modeled the setup in LTspiceXVII and it looks very promising.

I have laid out full-blown schematics and was halfway through building prototypes when I got moved to the UAE, so it was on hold a bit, but as there's allegedly a dearth of new ideas, I thought I'd throw this into the pot...

Possibly should also point out that I've considered pre-regulators and have modeled those too, both in linear and buck-converter SMPS form - the linear one uses an interesting TI ultra-low-noise LDO regulator and is in the V1 specification - the V2 may have the SMPS as an option, but it will depend a lot on whether I can manage the EMI. An interesting possibility if using an SMPS is configuring it as a buck/boost or just have a boost option, thus allowing higher voltage heaters from a lower voltage xfmr. Many possibilities.

More anon.

I'd be interested in folks' thoughts on this. :)

.
Multiple-PSUs-share-load.pdf
(186.43 KiB) Downloaded 247 times
Linear-ramp-generator-uses-one-op-amp.pdf
(106.04 KiB) Downloaded 267 times
Ramp trial.PNG
Last edited by jack on Wed Apr 12, 2017 4:09 pm, edited 4 times in total.
Vivitur ingenio, caetera mortis erunt
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#2 Re: My take on a heater supply...

Post by jack »

Seems that I can only add three attachments to a post, so here is an LTspiceXVII simulation run of the ramp gen - the output current of the board follows this precisely.
Ram Gen Sim.PNG
Vivitur ingenio, caetera mortis erunt
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#3 Re: My take on a heater supply...

Post by pre65 »

I'm not really very good (even poor) with surface mount parts, but a 10v @ 10A very low noise heater supply would be of interest.

I had been thinking of using a pair of 10v @5A Rod Coleman modules (which others have used for 833a) when I get round to finishing the transformer upgrades.

PS If a 10v @ 10A heater supply could be run off a 12v @30A SM power supply, all the better.
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#4 Re: My take on a heater supply...

Post by jack »

pre65 wrote: Wed Apr 12, 2017 3:29 pm I'm not really very good (even poor) with surface mount parts, but a 10v @ 10A very low noise heater supply would be of interest.
These could only be supplied pre-built - they cannot be conventionally soldered - an IR oven, SMT stencil and a lot of test kit is required to get this right.
Vivitur ingenio, caetera mortis erunt
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#5 Re: My take on a heater supply...

Post by pre65 »

jack wrote: Wed Apr 12, 2017 3:40 pm
pre65 wrote: Wed Apr 12, 2017 3:29 pm I'm not really very good (even poor) with surface mount parts, but a 10v @ 10A very low noise heater supply would be of interest.
These could only be supplied pre-built - they cannot be conventionally soldered - an IR oven, SMT stencil and a lot of test kit is required to get this right.
I see. If a pair of them (10v @ 10A) was comparable (not necessarily cheaper) to four of the Rod Coleman modules (10v @ 5A) then I would deffo be interested.
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#6 Re: My take on a heater supply...

Post by Mike H »

Image

Watching.

Just FYI the Rod Coleman uses a CCS on one side to get a soft start effect, the other polarity being a Voltage reg.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#7 Re: My take on a heater supply...

Post by jack »

Mike H wrote: Wed Apr 12, 2017 4:46 pm Image

Watching.

Just FYI the Rod Coleman uses a CCS on one side to get a soft start effect, the other polarity being a Voltage reg.
Rod's board is a CCS/Gyrator(aka voltage-controlled CCS) which is an interesting take - I've used these before and there's a good writeup by Ale at Bartola but I just like doing my own take on stuff... Ale also sells his own gyrator boards...

Didn't mention that the board has fold-back current limiting (if using the linear pre-regulator but could be implemented in the SMPS too), ULVO and digital (opto-isolated) on/off... The digital on/off allows you to do cute stuff, like sequencing startup and shutdown - the "shutting down" output allows you to chain to the HV counterpart (see below), so if the board detects a problem, you can automatically shut off the HV too... Also the possibility of adjustable ramp down on shutdown - would turn off the HV immediately then ramp the LT gently down...

I'm also interested in doing a complementary (not in the context of "free"!) high voltage, low(er) current board for anode and grid supplies...
Vivitur ingenio, caetera mortis erunt
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#8 Re: My take on a heater supply...

Post by ed »

Stirs a memory....Rod's boards are super quiet but very source hungry...IIRC puts big demands on heat sinking and you need agricultural caps to cope with any ripple current on the i/p side.....An alternative would interest me.
There's nowhere you can be that isn't where you're meant to be
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#9 Re: My take on a heater supply...

Post by Mike H »

Yes I used a couple for my 801A's, lo-otta heat! Even with big heatsinks. Even the toroid mains was roasting.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#10 Re: My take on a heater supply...

Post by jack »

This is the fundamental problem with linear pre-regulation of whatever type - you are dropping x Volts and drawing y Amps, hence needing to dissipate x*y Watts - with high current heaters, this can be a lot of energy.

The high current is also the issue with normal bridges - even with Schottky bridges with their lower Vf rating, you run into switching noise and the small choice in high-current Schottky devices - even when you locate a high If Schottky, as If increases, so does Vf - in the case of the 80SQ045N, as If approaches 8A, Vf is nearly 5.5V, i.e. the thermal saving is minimal.

Snubbing is a way to reduce switching noise (see the threads in which Nick & I commented on Morgan Jones' and others' work on this)...

With an Ideal diode bridge, most of these issues go away - there is no need to snub and the thermal loss is minimal - as FETs are used to switch and they are resistive devices, there is no Vf and losses at 50Hz are resistive only and low-RDSon, high Id FETs are common and cheap.

So, if Vin is higher than Vout, with a linear regulator you will dissipate approximately (Vin-Vout)*Iout Watts, i.e. consider this as 100% energy loss/

Therefore, a switching pre-regulator becomes far more interesting. Nothing new in this approach - switching pre-regulators have been used since the '70s and maybe earlier.

The world has moved on - SMPSs are now extremely efficient (90% is not uncommon), small, and the EMI issues are well understood - don;t forget that most mobile phones use SMPS supplies, as does a huge amount of industrial comms kit. Its not hard to get high current, ultra-low-noise SMPS buck & boost converters that produce only uV of noise. The EMI question has had to be addressed in order to get the appropriate certifications, so the most interesting combination from where I stand is an extremely high frequency, low-noise SMPS to do the bulk of the voltage adjustment, followed by an ultra-low-noise linear regulator.

The SMPS' voltage output is set to just above the linear regulator's Vout + Vdropout, where Vdropout is the regulator dropout voltage. This gives all the advantages of the linear regulator and true low-uVolt noise levels with the absolute minimum of thermal loss. It should be possible to reduce thermal losses by a factor of close to 10 vs. linear pre-regulation.

Nothing new in this approach - its been done for years in industry.
Vivitur ingenio, caetera mortis erunt
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#11 Re: My take on a heater supply...

Post by ed »

bring it on Jack..

I've just refreshed the suspect memory and I realise I went through 3 transformers and 2 sets of bridge/caps for the Coleman regs on my sv572. The transformers came from Simon Williams(BLS) in Durham and were a bit on the shy side...I realised at the end that the secondary current into the regs needs to be 3 times the heater requirement.....

so.... ended up feeding each reg with a 12v 12a smps...worked a treat..but it takes a lot of space and a lot of heat sinks..
so.... your solution would interest me and I might get the sv572 out of the cupboard, it was a cracking amp.
There's nowhere you can be that isn't where you're meant to be
User avatar
cressy
Shed dweller
Posts: 2906
Joined: Fri Aug 03, 2007 7:07 pm
Location: the great white space
Contact:

#12 Re: My take on a heater supply...

Post by cressy »

That reminds me ed, ive still got pauls pair of sv811-10s
I drew up a cct using a cathode follower to drive them then stalled with the heater supply

I was looking at shoving an smps straight up them bit then got sidetracked by the cold cathode resistance being a potential problem with the smps struggling to start the heaters. Did some reading, but i couldnt find any concensus as to wether it was an issue at all.

These type valves are a pig to heat, a board like nicks is ideal
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#13 Re: My take on a heater supply...

Post by jack »

Talk is cheap - this is a complex project I've had to pause half way through - i doubt that I'll have beta boards before the year end whilst the prototypes work, they are just parts of the solution, PoCs...

Putting it together reliably and designing a board that can take the current is an interesting challenge in itself.

Lots of other stuff to do with safety also need to be factored in.
Vivitur ingenio, caetera mortis erunt
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#14 Re: My take on a heater supply...

Post by IslandPink »

So is this a DHT heater ?
"Once you find out ... the Circumstances ; then you can go out"
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#15 Re: My take on a heater supply...

Post by jack »

IslandPink wrote: Thu Apr 13, 2017 10:15 pm So is this a DHT heater ?
DHT or normal indirect, yes
Vivitur ingenio, caetera mortis erunt
Post Reply