DHT Heater using VCCS - PCB or Kit (Possible Group buy)
#871 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Well, if we break the buy down into smaller chunks then you will forfeit the discounts, so April might be best.
Andrew
Andrew
Analogue, the lost world that lies between 0 and 1.
#872 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Andrew could I ask a few quick questions just to cement my understanding before I get the boards and start building and planning where things are going.
PRE-FILTERING
The topic with the support notes shows this...
2.5v = 10R (6.3v AC in)
... So I just wanted to confirm the TX connects directly to the board. No "raw DC" filtering required ?
I note some are using SMPS (laptop power supplies etc) as feed ins to the boards ? Is there an advantage to using either SMPS or pre-filtering before the board ?
In particular I'm planning to have a power supply box with umbilicals to the monoblock amps - if I don't pre-filter then I'll be running AC wires in the umbilicals which feels like a bad idea to me.
If I choose to pre-filter I can use PSUD to get the design right to supply the required voltage for the current draw. What DC voltage is required inbound to the board, and what current draw should I expect above that required for the filament ?
HEATSINKS
Could you or someone give me a quick understanding of how to calculate the size of heatsink I need ? So for example...
71A needs 5.0v @ 0.25A. 9v AC tx required, so we're dropping 4v in the reg = 1W dissipation.
45 needs 2.5v @ 1.5A. 6.3v AC tx, dropping 3.8v = 5.7W dissipation.
The heatsinks are rated for "deg/watt" right ? So that's how much the temperature rises based on the power being fed in ? Ventilation important to then lose that heat from the sink, but at what rate can it be expected to be lost ? How do you choose ?
Sorry if this is all basic stuff, just want to wrap my head around it before I plan dimensions and ventilation.
EDIT: I should not post until I've done my own research. I should not post until I've done my own research. I should not post until I've done my own research. I should not post until I've done my own research. I should not post until I've done my own research.
This looks like a good page for anyone with the same question: https://www.designworldonline.com/how-t ... heat-sink/
So simplifying that down:
Take the rating of the heatsink (say 3.4) and add approx 1 (transistor junction thermal resistance + insulating tab + grease)
Multiply that by the dissipation of the transistor : say 5.7 for the 45
Add ambient temperature : say 50 degC
gives me 50+5.7x(1+3.4) = 75 decC
I need to check the transistor attaching to the heatsink and see what it's de-rating is at 75 degC.
Do I have this about right ?
PRE-FILTERING
The topic with the support notes shows this...
2.5v = 10R (6.3v AC in)
... So I just wanted to confirm the TX connects directly to the board. No "raw DC" filtering required ?
I note some are using SMPS (laptop power supplies etc) as feed ins to the boards ? Is there an advantage to using either SMPS or pre-filtering before the board ?
In particular I'm planning to have a power supply box with umbilicals to the monoblock amps - if I don't pre-filter then I'll be running AC wires in the umbilicals which feels like a bad idea to me.
If I choose to pre-filter I can use PSUD to get the design right to supply the required voltage for the current draw. What DC voltage is required inbound to the board, and what current draw should I expect above that required for the filament ?
HEATSINKS
Could you or someone give me a quick understanding of how to calculate the size of heatsink I need ? So for example...
71A needs 5.0v @ 0.25A. 9v AC tx required, so we're dropping 4v in the reg = 1W dissipation.
45 needs 2.5v @ 1.5A. 6.3v AC tx, dropping 3.8v = 5.7W dissipation.
The heatsinks are rated for "deg/watt" right ? So that's how much the temperature rises based on the power being fed in ? Ventilation important to then lose that heat from the sink, but at what rate can it be expected to be lost ? How do you choose ?
Sorry if this is all basic stuff, just want to wrap my head around it before I plan dimensions and ventilation.
EDIT: I should not post until I've done my own research. I should not post until I've done my own research. I should not post until I've done my own research. I should not post until I've done my own research. I should not post until I've done my own research.
This looks like a good page for anyone with the same question: https://www.designworldonline.com/how-t ... heat-sink/
So simplifying that down:
Take the rating of the heatsink (say 3.4) and add approx 1 (transistor junction thermal resistance + insulating tab + grease)
Multiply that by the dissipation of the transistor : say 5.7 for the 45
Add ambient temperature : say 50 degC
gives me 50+5.7x(1+3.4) = 75 decC
I need to check the transistor attaching to the heatsink and see what it's de-rating is at 75 degC.
Do I have this about right ?
Last edited by RhythMick on Mon Mar 19, 2018 9:45 am, edited 1 time in total.
- IslandPink
- Amstrad Tower of Power
- Posts: 10041
- Joined: Tue May 29, 2007 7:01 pm
- Location: Denbigh, N.Wales
#873 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
I can answer a couple and let Andrew comment on the heatsinks.RhythMick wrote: ↑Mon Mar 19, 2018 8:52 am Andrew could I ask a few quick questions just to cement my understanding before I get the boards and start building and planning where things are going.
... So I just wanted to confirm the TX connects directly to the board. No "raw DC" filtering required ?
I note some are using SMPS (laptop power supplies etc) as feed ins to the boards ? Is there an advantage to using either SMPS or pre-filtering before the board ?
In particular I'm planning to have a power supply box with umbilicals to the monoblock amps - if I don't pre-filter then I'll be running AC wires in the umbilicals which feels like a bad idea to me.
Yes, you feed AC into these, no sub-board for rectification.
If you are concerned about hum, perhaps you can use some twin-core with braided outer for shielding, I have some in three core sizes which I use for AC feeds. It's white with Teflon jacket, I think I bought some originally for interconnects !
I used SMPS to generate 24V for GM70 and GK-71, mainly for cost reasons as you can get a suitable one for about £12, whereas the just the caps with suitable ripple current for this were looking like >£50 . Also more efficient.
"Once you find out ... the Circumstances ; then you can go out"
#874 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Much appreciated thanks. I'll keep it simple to start with and try either that or separate umbilical if the hum is bad. I guess better to mount the TX in the amp rather than the monoblock, so the low-current primary wires are in the umbilicals ?IslandPink wrote: ↑Mon Mar 19, 2018 9:14 am I can answer a couple and let Andrew comment on the heatsinks.
Yes, you feed AC into these, no sub-board for rectification.
If you are concerned about hum, perhaps you can use some twin-core with braided outer for shielding, I have some in three core sizes which I use for AC feeds. It's white with Teflon jacket, I think I bought some originally for interconnects !
I used SMPS to generate 24V for GM70 and GK-71, mainly for cost reasons as you can get a suitable one for about £12, whereas the just the caps with suitable ripple current for this were looking like >£50 . Also more efficient.
What's the typical additional current draw from the board, or is the current draw as per the filament and the voltage is higher ? If this is correct then
26 : 6.3v @ 1.05A = 6.6VA = 27mA primary
71A : 9v @ 0.25A = 2.25VA = 9mA primary
45 : 6.3v @ 1.5A = 9.45VA = 39mA primary
each valve x 2 so roughly 150mA in the primary wires @ 240v. That's probably worth shielding, or pre-filtering.
#876 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
What is the transistor attaching to the heatsink please ? I'd like to have a look at the datasheet so I understand the thermal requirements.RhythMick wrote: ↑Mon Mar 19, 2018 8:52 am This looks like a good page for anyone with the same question: https://www.designworldonline.com/how-t ... heat-sink/
So simplifying that down:
Take the rating of the heatsink (say 3.4) and add approx 1 (transistor junction thermal resistance + insulating tab + grease)
Multiply that by the dissipation of the transistor : say 5.7 for the 45
Add ambient temperature : say 50 degC
gives me 50+5.7x(1+3.4) = 75 decC
I need to check the transistor attaching to the heatsink and see what it's de-rating is at 75 degC.
Do I have this about right ?
Nevertheless, it looks to me like the 71A is no problem and could use the smallest heatsink available, but the 26 and 45 need more attention. With 50 degC ambient and a 4.5 degC/W heatsink I'm looking at a junction temperature of around 80 degC I think.
#877 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Hi, on the small boards the heatsink is shared between the LD1085 and the D44H11.
Th extra current drawn by the board is inconsequential compared to the DHT.
You will need to ensure the correct AC by choosing a TX with suitable secondaries, anything over and above requirements will place extra load on the pre regulator.
Andrew
Th extra current drawn by the board is inconsequential compared to the DHT.
You will need to ensure the correct AC by choosing a TX with suitable secondaries, anything over and above requirements will place extra load on the pre regulator.
Andrew
Last edited by Andrew on Mon Mar 19, 2018 1:22 pm, edited 2 times in total.
Analogue, the lost world that lies between 0 and 1.
#878 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Thomas +4
Andrew
Andrew
Analogue, the lost world that lies between 0 and 1.
#879 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
RhythMick wrote: ↑Mon Mar 19, 2018 12:55 pmWhat is the transistor attaching to the heatsink please ? I'd like to have a look at the datasheet so I understand the thermal requirements.RhythMick wrote: ↑Mon Mar 19, 2018 8:52 am This looks like a good page for anyone with the same question: https://www.designworldonline.com/how-t ... heat-sink/
So simplifying that down:
Take the rating of the heatsink (say 3.4) and add approx 1 (transistor junction thermal resistance + insulating tab + grease)
Multiply that by the dissipation of the transistor : say 5.7 for the 45
Add ambient temperature : say 50 degC
gives me 50+5.7x(1+3.4) = 75 decC
I need to check the transistor attaching to the heatsink and see what it's de-rating is at 75 degC.
Do I have this about right ?
Nevertheless, it looks to me like the 71A is no problem and could use the smallest heatsink available, but the 26 and 45 need more attention. With 50 degC ambient and a 4.5 degC/W heatsink I'm looking at a junction temperature of around 80 degC I think.
I'm pursuing this thoroughly because I need to consider how and where I'm mounting these boards and the ventilation needed around them.
It would be good if I can use for example the 25.4mm heatsink, which is 7.8 degC/W, but that implies a junction temperature of 100 degC. Within the operating range of the LD1085V50 which I found referenced earlier in the thread (-40 to +125 degC), but feels like it's pushing it a bit. I'll have 4 of these (plus 2 of the 71A version) in a space under the valves which is around 350mm x 160mm and this is going to get a bit toasty I think.
http://uk.farnell.com/search?st=SK129
#880 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Many people are using 300B, the heaters dissipate 5v at 1.2A they run them from 9v AC secondaries, call it 5 watts to be dissipated, the 50mm heatsink is usually fine, that's what I have on my bread boarded setup but the 63m is probably better.
If I recall, I did some thermal testing and posted the results this thread.
Andrew
If I recall, I did some thermal testing and posted the results this thread.
Andrew
Analogue, the lost world that lies between 0 and 1.
#881 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Thanks Andrew. You'd think having read all 57 pages TWICE I'd have spotted that
I think I've got my head around it a bit more now.
What size heatsink is in the kit?
Am I right in saying it basically needs about 4v headroom to regulate? So for
26 (1.5v) use 6. 3v tx
71A (5v) use 9v tx
45 (2.5v) use 6.3v? Is 3.8v enough headroom?
I think I've got my head around it a bit more now.
What size heatsink is in the kit?
Am I right in saying it basically needs about 4v headroom to regulate? So for
26 (1.5v) use 6. 3v tx
71A (5v) use 9v tx
45 (2.5v) use 6.3v? Is 3.8v enough headroom?
#882 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
I can do any of the sizes.
Yes, roughly 4v. Are you taking into account of the fact you need to multiple the AC by 1.41 to get the raw DC?
Andrew
Analogue, the lost world that lies between 0 and 1.
#883 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
D'oh. I'll get my coat.
Having thought through the location I may as well put the larger heatsinks on as they'll all be at the same height. I take it you'll need to know before you ship.
#884 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
My final build will be a power smoothing and distribution box with an umbilical to the stereo amp. For the DC heater boards I have two options...
1) feed 240v AC mains into the amp with 1 tx per DC board in the amp - total mains draw for heaters would be around 250mA
2) pre-reg the mains in the power box and feed the DC boards with separate wires in the umbilical
I'm seriously considering approach 2.
PROS
- no AC in the umbilical or amp
- less components in the amp
- unlikely to improve the already excellent output from the board, but certainly shouldn't harm it
CONS
- more wires in the umbilical
- more components in the power box
Keen to get views from the forum.
1) feed 240v AC mains into the amp with 1 tx per DC board in the amp - total mains draw for heaters would be around 250mA
2) pre-reg the mains in the power box and feed the DC boards with separate wires in the umbilical
I'm seriously considering approach 2.
PROS
- no AC in the umbilical or amp
- less components in the amp
- unlikely to improve the already excellent output from the board, but certainly shouldn't harm it
CONS
- more wires in the umbilical
- more components in the power box
Keen to get views from the forum.
#885 Re: DHT Heater using VCCS - PCB or Kit (Possible Group buy)
Andrew will say for certain, but there is no reason why you cant feed the boards DC if you want to do the initial rectification remotely. But consider you are connecting the cathode of your valves to a long wire going back to the second box just to have two boxes. TBH, dont underestimate the problems that two boxes can cause.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.