Starting added tube 50w using Marstran chassis & iron

A place to discuss wiring, installation, diagrams and so on.

Starting added tube 50w using Marstran chassis & iron

Postby ivan H » Sat May 04, 2013 1:18 am

Hi every1, I've been wanting to do an added tube (Caswell #39) type build for a while & have finally got started. I had hoped to use the 50watt trem/mod chassis that Brian was trying to get happening, but when this didnt come to fruition Brian kindly offered to drill 1 o his existing chassis for an extra pre amp tube, thank you very much Brian. Very nice chassis too. The PT is Brians 1202-118 & the OT I chose is Brians custom 50 watt Dagnall clone. Choke is Brians 3h C1999. I am waiting on the last order from Robert at Valvestorm to arrive which contains the front & rear panels, ARS eletrolytics, pots & a few other components so I'll have to wait until then to get right into it. As I dont need it to be period correct or anything I'm using a Metro Amp plexi front panel & one of Sparky's rear panels. So far I have the transformers & choke mounted along with ceramic valve sockets. All pre amp sokets have been rotated 180 degrees so the input goes to V1b first, as this is supposed to be the "quieter" side of the tube, & all components are then all laid out in the correct order on the board without the signal crossing over earlier stages anywhere. V2 is only 1/2 heatered with V2b's heater being connected only. Brian added the socket hole between V2 & V3 which fits in perfectly with the board layout with only the two 10k pre amp HT voltage droppers/decouplers being arranged differently to make it all fit neatly. These two 10k resistors I have arranged "in line" rather than side by side. I also "squeezed" the tone stack slope resistor down by 1 turret space (or hole space) on the board. These 2 changes freed up 2 rows of holes on the board so now components for the last gain stage/cathode follower, which has been moved are in the correct place relative to the tube. For heater wiring to the power tubes I have used solid 18g green cloth covered heater hookup wire, then the standard 22g red & black wire for all pre's. The heater CT is connected to position 1 of Larry's grounding scheme with the solid 18g wire. For the turret board I got a blank piece of Brian Haberman's nice board material & have populated it with turrets myself. It is not that different from a stock 50 watt board. I have Sozo coupling caps & most of the resistor will be a mix of 1/2w & 1w Vishay Dale metal films (for a low noise floor), with a few carbon films. Voltage droppers/decouplers are 3w metal oxides. I have made up the input hot shield by using the normal green 22g wire (from Valvestorm's plexi wiring kit) which I inserted into a 4" piece of braided sleeve & then covered with double walled heatshrink. The board is wired & partially populated with components. I have pics so far & will keep a good record of the build with pics that I will have to take to a friend with a computer to upload (as I'm computer illiterate). I would very much appreciate any tips from any 1 who has completed this type of build before. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Advertisement

Banner ads are available for $25 a month. PM or Email me at Info@wallaceamps.com for more details.

Re: Starting added tube 50w using Marstran chassis & iron

Postby neikeel » Mon May 06, 2013 4:37 pm

Good stuff!

I have been tweaking my SIR 39 clone in snatches of time recently. I set it out to use V1b onwards as stock 69 spec lead amp, including the 68k grids on the board, 0.68uF on V2a 0.0022uF coupler and 47k/8ohm NFB.

For V0 I used an extra board with the dedicated input jack in the hi normal channel hole, V0b as 3k/0.68uF via a 0.022uF coupler and 470k/2200pF peaker onto the gain pot, then back to V0a grid then 0.022uF coupler, this stage is 8k2 bypassed by a 0.22uF cap, to a push-pull switch on the back of the gain pot (mounted in the lo normal channel hole). Using Vol 2 pot as a pre-pi mv (already have PPIMV in the trem speed hole). In 2203 mode the wire from the switch goes to an unbypassed 470k mixer.

When you pull the pot out the wire goes into a 470k/470pF/470k voltage divider and then to V1a which has a 10k unbypassed cathode resistor and 0.022uF coupler. The 100k plate resistor has 500pF ceramic over it.

Works pretty well with 1000pF bright cap on the 2203/4stage gain pot and 150k to ground on the pot too.

It has plenty of gain (I could get more if I lowered the cathode resistors) for me and plenty of low end too. I do need to implement the Larry grounding tho' :wink:
Neil
neikeel
 
Posts: 268
Joined: Wed Jun 25, 2008 11:24 am

Re: Starting added tube 50w using Marstran chassis & iron

Postby Did J » Tue May 07, 2013 1:58 pm

It seems you've got la vista! Nice report on your projects. You're very talented :wink:
Did J
 
Posts: 16
Joined: Wed Nov 17, 2010 6:04 pm

Postby ivan H » Tue May 14, 2013 5:21 am

Hi again Neil, I'll take a few more pointers off your post above. The 2200pf peaker over the 470k before gain 1 I was planning on using, I read about in the SIR 100 watt superlead thread on the Metro forum, along with a lot of other good info. Im not sure about using a .0022 coupler,,, tho some reccomend it after the added stage so that be part of the tweaking process. I'm thinking with the coupler out of the first 10k stage I'll put thru an unbypassed 470k & then maybe try putting straight to the wiper of gain 2, just to see how it goes. I have 33k grid stoppers on the first 3 stages. I'll only add plate bypasses to any stages giving high frequency problems & start small, increasing the value as needed. I'll keep u guys informed. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Starting added tube 50w using Marstran chassis & iron

Postby ivan H » Sun May 19, 2013 8:52 pm

Ok, the last package from Valvestorm arrived a bit over a week ago so I've been doing a bit each evening. I'm really impressed with the rear panel from Sparky & think I'll order a properly marked 1 once I've tweaked it to my liking. I have drilled between the Hi & Lo number 1 inputs to add the gain 1 pot . All chassis wiring is complete (using Larry's grounding scheme) except for the 240VAC wiring & indicator lamp. I had hoped to have this done & the board, which is completely wired & populated, installed over the weekend but a few other jobs needed doing so I didnt have time. Circuit wise I have decided to start with a #39 mode similar to the spec Neil posted a few days ago in the SIR 100 watt SL thread on the Metro forum, the only difference being the first stage is the standard 2k7/.68uf. This stage is coupled by an .022uf to a 470k/.0022 peaker to the 1 meg pot which has a 150k from the wiper to earth. This goes thru a 33k grid stopper mounted on the socket of the first cold biased stage. This 10k cathode resistor is bypassed with a .22uf in #39 mode. This will go to the normally closed terminal of 1 side of a DPDT Omron relay. The same cathode resistor will be bypassed with a .1uf in #36 mode, connected to the normally open terminal of the relay. Both caps have a 100k resistor connecting them to earth, with the relay bridging the 100k resistor on the selected cap. This stage is coupled with a .022uf into an unbypassed 470k resistance to the gain 2 pot. Again there is a 33k grid stopper on the 2nd cold biased stage. This 10k cathode resistor is not bypassed. In #36 mode it has another 10k switched in parallel by the DPDT relay. Again the coupler is a .022uf which goes to a 470k/500pf/470k divider then to the 820 ohm/cathode follower stage. This is the spec I'll use as a starting point. I'm expecting it'll need quite a bit of tweaking, but thats half the fun, isnt it. During tweaking I'll decide on whether to use a switchable 220/470pf tone stack or not & also whether to add a switchable .68uf/unbypassed/25uf to the 820 ohm cathode resistor before the cathode follower. There is 100k series resistor between the pre PI master volume & the PI. NFB is 47k/8 ohm with a .0047uf/1meg resonance control & 5k/.68uf presence. Hopefully I get the board in over the next couple of days. I'll try to get a couple of pics up here. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Starting added tube 50w using Marstran chassis & iron

Postby ivan H » Sat May 25, 2013 11:59 pm

Hi again. Ok so today I am putting the board in. I havent had as much time to spend on this as I'd like but not to worry. I have decided not to bother with the switchable .68uf/unbypassed/25uf on the 820 ohm cathode b4 the cathode follower as it would most likely make it too gainy/unruly, & I have also just put in a 500pf treble cap in the tone stack for the time being, rather than a switchable 220/470pf. I can always add this later if I want to. I have built up the power supply for the cathode bypass switching relay on a tag strip mounted on the chassis sidewall beside the PT. This is powered off the 5V winding on the PT. I used the same supply that Ceriatone use on their AFD35's, which is a voltage doubler type rectifier that outputs to a 7805 +5V regulator. This is also mounted on the chassis sidewall. I have run a piece of solid 18g heater hookup wire under the pots as an earth bus (I know its way overkill but I've got it laying around) for the footswitching system that incorporates a DPDT switch, indicator LED & jack socket for the footswitch. Whether its better to do this bus or use the chassis for the earth points I dont know. The relay & LED + leads are inserted into a piece of heatshrink & also run under the pots. The DPDT relay (Omron) is mounted upside down on a small piece of vero board that is mounted on standoffs between the PTP board & the 2 gain pots. This fits in neatly with Larry's grounding scheme as 1 pole of the relay selects either a .22uf or .1uf bypass over the 1st 10k cathode by shorting either of the two 100k resistors that tie the caps to earth & as such, this pole must earth at the 1st earth point. The other pole on the relay brings another 10k resistor in parallel with the 2nd 10k cathode stage & so must earth at the 2nd earth point. This arrangment gives a #39 ish mode with an added .22uf over the 1st 10k cathode, switchable to a #36 ish mode with a .1uf over the 1st 10k cathode & an unbypassed 5k cathode on the added stage. I have asked a friend to assist in getting a few pics up here later in the week (as I'm totally computer illiterate), so hopefully that happens b4 I have more to post about. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Re: Starting added tube 50w using Marstran chassis & iron

Postby neikeel » Sun May 26, 2013 4:04 pm

8) (shame no popcorn icon here!)
Neil
neikeel
 
Posts: 268
Joined: Wed Jun 25, 2008 11:24 am

Re: Starting added tube 50w using Marstran chassis & iron

Postby Brian Wallace » Sun May 26, 2013 8:58 pm

I should get more smilies.
Remember....You can fix it with a little extra energy and a positive attitude. ;)
User avatar
Brian Wallace
Site Admin
 
Posts: 1087
Joined: Fri Jun 06, 2008 7:34 pm
Location: Livonia, Michigan

Starting added tube 50w using Marstran chassis & iron

Postby ivan H » Mon May 27, 2013 6:59 am

Hi, I'll jst quickly add a few point that I forgot to include in my last post. I have added a resonance control (1 meg/.0047uf)on the rear panel, on the opposite side of the impedance selector than the speaker outputs. I have also added a PPIMV (dual ganged 250k pot/2.2 meg carbon film resistors) in the furthest speaker output hole from the impedance selector. The shielded leads from the PI output couplers to the PPIMV were made by twisting the green & orange wires & then inserting them into a piece of braided sleeve to which I had attached a short lead. I then insulated it by inserting into double walled heatshrink. (I also added another layer of this larger diameter heatshrink to the input hot shield lead I had earlier made, to be sure to be sure). If I were to use this amp only for gigging I probably wouldnt put the PPIMV in, but to use this at home it nescessary. The NFB resistor at the moment is a 47k carbon film (8 ohm tap), tho I'm considering dropping this to 22k & putting it in series with a 25k linear pot mounted beside the resonance control, so that when the PPIMV is turned down, the NFB resistor value can also be decreased in an attempt to keep the presence & resonance controls active. I'll cross this bridge while tweaking. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Re: Starting added tube 50w using Marstran chassis & iron

Postby neikeel » Mon May 27, 2013 6:48 pm

ivan H wrote: The NFB resistor at the moment is a 47k carbon film (8 ohm tap), tho I'm considering dropping this to 22k & putting it in series with a 25k linear pot mounted beside the resonance control, so that when the PPIMV is turned down, the NFB resistor value can also be decreased in an attempt to keep the presence & resonance controls active. I'll cross this bridge while tweaking. Cheers


Sounds like a good idea, I find I like my resonance control all the way up so will fix things like that and swap it for a variable NFB. On mine you really need some way of taking the volume down in cascaded mode as it sounds best with the premaster at least over 1/2.
Neil
neikeel
 
Posts: 268
Joined: Wed Jun 25, 2008 11:24 am

Starting added tube 50w using Marstran chassis & iron

Postby ivan H » Mon May 27, 2013 8:10 pm

Thanks for the tips Neil, u've been a wealth of information with this project & I appreciate it. I'd really like to do a 100 watt build but realisticly, I'd never use it, at least not without an attenuator, (which I dont have). For pubs & gigs of that size a 100 watter is just too much. Even a 50 watt,,, uno. I think for my next project I'll do something smaller that is also more versatile. Something like an 18 watt version of ur plexi/2204/#39 would fit the bill nicely. But that down the track a bit. For now I'd best not miss my bus to the daily grind. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Postby ivan H » Tue Jun 04, 2013 2:57 am

Hi, sorry I've been absent for a while, I was laid up for a few days with a terrible flu thanks to getting caught in the rain a couple of days b4 the start of winter. I was hoping pics would be up but have arranged to go to my friends & try posting some tomorrow. Anyway I have the amp all together & have just temporarily installed a 100k current limiting resistor between the rectifier & the main filter & powered it up with my old fluke 11 series multimeter hooked to read the voltage drop across the resistor. On power up the meter registered 274V, rapidly dropping. After 1 minute the meter registered 70V. After 3 minutes, 26V. After 10 minutes its registering 14V, & after another 5 minutes its registering 9V, so these ARS electrolytics from Valvestorm are indeed reformed not so long ago as claimed, thank you very much Robert. I imagine they would be fine to use as they arrive from Valvestorm. Already the meter is reading a 7V drop across the resistor so I expect that 1/2 hour will be all that is needed to get the voltage drop below 5V. As yet I havent ordered any valves, but they arrive within t days so its not a problem. I was thinking of using 6550's but have decided to stick with EL34's. I've always liked them so will order winged C's again. I have the ground leads from position 5 (Larry's scheme) to both power tube sockets. They are twisted as I figure they carry current alternately, whether twisting is needed I'm not sure. The leads connect to pin 1 (suppressor grids) with a 2 watt 1 ohm metal film bias sensing resistor wired between pins 1 & 8 (cathodes). So, the meter is reading 5.12V now so thats all good. I'll give it a while longer then check the bias range & other voltages throughout. For the phase inverter I have a Svetlana 12AX7 that my tube supplier has previously sent me for that use as it had well balanced output. For the cathode follower I'll order a Tung Sol ECC803S & most probably a couple of Mullard reissues for the first two positions, unless someone can reccomend I try something else in these 2. Circuit wise I have 34k grid stoppers on the first 3 stages. All coupling caps are 0.022uf. The treble peaker b4 the first gain control is 470k/0.0022uf. I used a sozo for this with the outter foil marker orient ed towards the pot. There is no treble peaker b4 the 2nd gain control, just a 470k resistor that I have put straight to the pots wiper. The pot is 1 meg at the moment tho I'll change it to 500k if I think it needs. As yet I have no anode bypass caps installed, I'll add them where needed. For these I have a few each of 100, 220 & 470pf 1000V ceramic disk caps. After the 3rd (added) stage I have a 470k/470k/500pf voltage divider/peaker, tho the divider is another area I'll be keeping an eye for change on. So, hopefully I'll get some pics up tomorrow. The voltage drop across the 100k resistor is down to 4.86V so all good. I'll just quickly add this edit, with the 100k current limiter removed & the rectifier correctly connected to the main filter the voltage is steady at 462V. At the screen filter there is 458V, PI filter has 457V, as does the cathode follower plate. The 820 cathode stage b4 the cathode follower has 452 at the plate, as does the added stage. The 1st & 2nd stages have 451V at the plates so with tubes in all should be in the right range. Bias with a 150k AC dropper installed sweeps from -39V to -54V. The output from the +5V regulator is correct & the relay is switching/indicator indicating off both the panel mounted switch & the footswitch. So I'll order some valves & get into tweaking. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Postby ivan H » Fri Jun 07, 2013 8:20 am

Ok, I will try and get these pics up!


Image


Pre amp sockets rotated 180 degrees so input goes to V1b. V2 is only 1/2 heatered with the b triode being used. The hotshield can be seen with the shield lead made to connect to the turret, not the socket pin.


Image


IEC connector's active is wired to tip of the fuse holder. The PT primary leads are secured by insulated terminal strip.Twisted brown & blue wires are 240VAC, twisted yellow wires are the 5VAC to the relay PSU. Twisted red wires are PT secondries to the standby. Relay PSU is on the sidewall beside the PT.


Image


Relay mounted upside down on small vero board which is on stand offs. Panel mounted switch & LED #36 indicator between gain 2 & pre PI master. Footswitch socket is the upper in pic (lower when amp is upright). U can see the ground lead from the near side of the relay's board (pole for switching 1st 10k cathodes bypass caps)
tied to the 1st position of Larry's ground scheme. The relay's other pole (for switching another 10k in parallel with the added tubes 10k cathode resistor) will be grounded at position 2 of Larry's scheme.

Image


Pre PI master, tone stack & presence pots partially wired (except connections to board). Ground leads are tucked behind pots to their respective ground points. Presence is 5k/.68uf. Heater CT & relay ground bus plus the relay & LED power leads which are enclosed in double walled heatshrink can be seen under the pots. They are tucked behind the pot ground leads to keep them secure.


Image


The speaker jack is a cliff switching type. I later connected the switch tag to the common tags to protect Brians custom OT should the amp ever be powered up with out the speaker load connected. OT secondary leads were zip tied close to the stand off before the board went in. This keeps them in the same position under the board as when they are run between the power tubes (under the bias network/voltage droppers) but avoids running them parallel to the high current power tube heater wiring. The resonance control is 1Meg/0.0047uf.


Image


Conditioning filter caps (first power up, no valves in). Multimeter leads can be seen draped over right side of chassis. These are hooked across a 100k current limiting resistor temporarily installed between the rectifier & main filter. No plate bypass caps are installed here, thou I have since added 1000V ceramic disk caps, a 100pf acrossthe 2nd stage plate resistor & a 220pf across the 3rd stage.


Image


The treble peaker b4 gain 1 is made up of a 470k/.0022uf Sozo. The outter foil marker is towards the pot. This peaker is mounted between the board & pot. No peaker b4 gain 2, just a 470k, also mounted between board & pot. I used a 100k between the pre PI master & the PI, also mounted between board & pot (can just be seen at left of pic). A blue 1/2 watt 150k can be seen between the wiper & ground lugs of the gain 1 pot & two of the same type 100k's tying the 1st 10k cathode's bypass caps to ground (near the input jack socket). These will later be replaced with better quality Vishay Dale's.


Image


Front panel before knobs fitted. This will probably be replaced with a correctly labelled panel from sparky.


Image


The resonance control will probably be moved towards the HT fuse & a variable NFB pot (25k liner in series with a 22k on the board) put in its place. The amp will be properly tidied up when I have tweaked to my liking. I'll use an A/B switch to switch between my switchable rectifier/main filter 50 watter & this amp.

Cheers...
Last edited by ivan H on Sat Jun 08, 2013 1:54 am, edited 9 times in total.
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Re: Starting added tube 50w using Marstran chassis & iron

Postby neikeel » Fri Jun 07, 2013 10:20 am

More, more (esp foot switching stuff!) 8)
Neil
neikeel
 
Posts: 268
Joined: Wed Jun 25, 2008 11:24 am

Postby ivan H » Fri Jun 07, 2013 7:52 pm

Thanks Neil, I'll see if I can get a couple of close ups of the relay PSU & the DPDT switch/LED network up here for u. Schematically, I lifted it straight from the Ceriatone AFD35 layout, I just changed the layout to put it on tagstrip. Sorry about the pic quality, I only have my phone to take them with. I'd like to thank my friend for taking the time to get these up for me. I'd also very much like to thank Brian Wallace here at Marstran for the very nice chassis & iron, Robert at Valvestorm for supplying the bulk of the components, Brian Haberman for the nice board material (its looks heaps better & is so much easier to use than G10), Sparky & Metro Amp for the panels. Without guys like you, projects like this wouldnt be possible, so thank you all very much. I have 2 winged C EL34's & a Tung Sol ECC803S for the cathode follower on the way. I already have a Svetlana 12AX7 with matched outputs for the PI. I'm still undecided on wot 2 use for the 1st & 2nd pre's, I'll go thru a few old pulls to try finding wot sounds best. I'd like to try the Genalex Gold Lion ECC83, they're just so expensive. Cheers
ivan H
 
Posts: 35
Joined: Thu Dec 27, 2012 4:49 am

Next

Return to Marstran Transformers

Who is online

Users browsing this forum: No registered users and 1 guest


cron