'SE Heaven' has/had this intermittant annoying fault of arcing his likkle 6EM7's on power up. The result is a few seconds of fixzzlepoppensparken through the speakers, it can't be good for my lovely speakers or the yummy biscuit base RCA NOS 6EM7's I've been using.
I want to do something to him, a new lunchtime project.
The current PSU is thus...
Now, I've never built anything with a Valve rectifier in and I'm wondering if that would be the way to go on 'SE Heaven'? A slow startup? I'm willing to spend out on a new mains TX (I don't have a dedicated mains TX with a 5volt heater supply but might be able to 'find' a SMPS 5V PSU).
I like the way 6EM7 sounds through MLTL's so want to keep him (for now), any advice? Is valve rec the way to go here? I dunno.
Hi Dave-you could try a thermistor to give the transformer less of a surge on switch on (very cheap mod !) or if you want to be sophisticated like what me and Andrew I are you could use a couple of 6D22S television damper diodes.
These also give a slow warm up (bit longer than GZ34/37) and have the added advantage (?) of 6.3v heater supplies.
Why not try a thermistor first ?
PS
I used a 4R 8 amp Rapid pt no 26-3326 but they don't seem to list that one anymore !! (called a current inrush limiter)
Attachments
Last edited by pre65 on Mon Mar 10, 2008 11:26 am, edited 1 time in total.
The only thing necessary for the triumph of evil is for good men to do nothing.
Cheers for the input Phil but I think my problem is that the SS HT is too high (off load) at the point just before the valve starts conducting. Hence the arcing.
I dunno if an in-rush filter would help things there would they?
The 6D22S are one of a few rectifryers I could use I suppose.
I tried implementing Wills SS clamping diodes but just managed to turn one into a short cct a while back
AS you say day, its the period before teh 6em7 is conducting.
The problem with the thermistor is it will just stop the surge, it may not delay long enough.
The problem is that the 350v TX will be producting about 480v when the 6em7 are not drawing current (350 * 1.4) and this will drop to 320 when its pulling current (350 * 0.9).
One way would be to add a dropper to ensure the current at power on is above critical value, so its always running in choke input. About 45ma should do the job with a 10H choke. but if (As the diag shows) the choke can only stand 100ma, its a no go, as the total constant would be 150ma or so.
So if you used a rectifier, you need to make sure its a slow startup one, or the B+ will get there before the 6em7 is conducting. If you were thinking about using a SMPS, check it sok with its output a B+ above ground.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Cheers Nick, the 10H choke happily copes with the demand of the 2 6EM7's so would I be better off just getting a mains TX better suited to the 320V rail I'm aiming at.
Re- the SMPS, I was thinking of using just one to power the rectifier heater element, would that need to be ref'd to the rest of the cct? I guessed it wouldn't hence my suggestion as it's a cheap way of 'going valve rect'. They drop approx 70V too don't they?
The SMPS might be ok, but it depends, obviously if you use a directly heated rectifier, it will be at B+, and I would check what the max heater/cathode voltage for indirectly heated rects. the damper diodes Phil suggested would do the job, and they have nice slow startups.
You could change the TX, but its actually a good TX, and choke input is good, as long as you can manage the startup. You will normally get better regulation with choke input than cap input.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Hmmm, OK then, as Phil pointed out they have 6.3V heater requirements too which means I could press into service an old Mapperlins TX for 6D22 heaters. I think I'm pushing the heater windings on the 0-350 mains TX as it's supplying the 2 6EM7's and it's only rated at 2Amps so I'll need to sort that anyway.
EDIT. They're halfwave diodes aren't they too I just realised. The TX hasn't got a CT so ASFAIK I can't go FW rect (2x 6D22's). I'm back to a bridge aren't I? 4 x 6D22s's??? It's been a long day. Help!
Of course the old and trusty GZ34 is a slow start full wave rectumfriar but you don't have to use 6D22S although there are loads of these waiting for you, in Ukraine and Lithuania and Russia.
Consider the unlooked for PY500A. It has 40volt 300ma so nice and low current heaters and is otherwise an English/German version of the 6D22S. You see 6D22s use I think 2amps per heater and as they are half wave rects, you need two, so 4amps. If you have loads of 6.3volt amps then fine, otherwise consider something else!
I've edited my PSU piccie as it was too big, sorry, made the page hard to read.
Right, seeing as each 6D22S is going to pull 2A the maplins PSU will keel over so I'll need to look for another bumfryer heater TX possibly.
GZ34 is a full wave as Mr I says so only need one but can I do FW rect on a TX without a CT? I thought a bridge was the only way? Thats what I've done up to now.
I need to go home, brain is zapped today. (New video conferencing system lessons today, eeerrrrrccckkk).
1) The old double switcher-roonie, first flick fires up the heaters, count 1.2.3.4...10, flick second switch to fire up HT, I think Philip has a nice circuit somewhere that allows you to wire up the switches so it doesn't matter which one is flicked first, the heaters always come on....then the HT
2) A time delay relay, same as the above but automated. You could build one with a 555 timer, a darlington pair and a normal relay, if you prefer.
That's a bit luxurious and expensive looking, Nick!
The only problem I encounter with having the heaters warm through first is that when I flick the HT switch there is a thump that goes through the circuit - not aesthetically pleasing.
