Amp rebuilt with iskras & mustards

ivan H

Ambassador of Tubes & Grooves
Country flag
I first built my main amp, a 4 hole 50 watt m@rsh@ll clone around 11 yrs ago, using commonly available carbon film resistors & Mallory 150 series metalized polyester film capacitors. Tweaked to a spec I was happy with using features from the '67 & '68 50 watt amps, it has served me well & sounded very good. Over the years however, I often thought about a build using more "period correct" components, so started slowly gathering components.
Now I must say a very big "thank you" & "very much appreciated" to one Mr RVA for his assistance in acquiring a complete set of period correct Iskra resistors, without which this project would still be in the "parts aquisition" stage, so thank you very much Ray, you are a scholar & a gentleman. So, with parts acquired the amp was stripped to thisWP_20181221_09_00_21_Smart.jpg
Cleaned up & new ceramic pre amp tube tube sockets installed (to replace micalex ones), the board cleaned up & re turreted & rewired, then installedWP_20181221_14_53_55_Smart.jpg
& then populatedWP_20181222_13_44_54_Smart.jpg The C296 mustard caps are mostly Philips Mullards, the exceptions being the 0.0022uf V1B coupling cap, which is an Aussie made Philips Miniwatt, made at the Hendon SA plant, & the two 0.68uf/160V V1B & V2A cathode bypass caps, which are Spanish Philips Bianchi types. These "foreign" mustards are made on the same machine types as the Mullards, with the same material stock supplied by parent company Philips, so while never used in Marshall's, they're the same cap type. Not what you'd use in a vintage Plexi restoration but plenty good for clone amps. The HT voltage dropping resistors are period correct 1 watt RS morganites. The treble & treble peaking caps are 60's era RBS 2% 560pf silver mica's. I'm still looking for two RS types for these positions. Volume 1 bright cap is a 4700pf/1000V Lemco ceramic disk, though this may well be changed to a smaller value.
After going through all checks I installed the same valves that had previously been in it, power tubes biased to 66%.
So how does it compare? Well it's noticeably different & to my ears for the better. Much smoother, mids are more open. Seems richer with better harmonic content & has that "note bloom" thing happening much more than it did before. After putting about an hour on it I changed out the tubes for the new ones I had for. Rectifier is a NOS/NIB Valvo branded f32 Blackburn Mullards GZ34. EL34's are the same (St Petersburg) Winged C's that have been in it. Of the two NOS Brimar CV4004's that I had for it, the first that I tried in the phase inverter position gave a plate differential voltage of close to 11v, so it was relegated to the V2 spot. The other Brimar in the phase inverter socket gives a plate differential voltage of 7.8v, a little higher than I'd like but still in the ballpark & will do for now. V1 is a NOS Mullard 163 series 12AX7. I've put around 5 hours on the amp with these tubes in & really like the grit of the 163 Mullard in V1, however it is just a little noisey (not microphone, just a tad noisey on the bright channel). I do have a NOS/NIB factory matched duet of Mullard XF2 EL34's & another 163 series 12AX7 on the way so plan on doing some tube rolling & maybe a couple of small circuit tweaks as the amp settles in & will then try to get a couple of clips up. A couple more picsWP_20181225_08_23_03_Smart.jpg
WP_20181225_08_21_14_Smart.jpg
Cheers
 
Last edited:
Mr Cheers,,,,,,,,,,,,,,,, That sure is something. Now I get to wondering a few things. Do you Aussies get powered at 240V or 120?
Can your amp do both? My '72 does since it was made in England obviously, then used in Germany and eventually into my hands here in America.

PS That sure is some tidy work there, Ivan.
 
Last edited:
I first built my main amp, a 4 hole 50 watt m@rsh@ll clone around 11 yrs ago, using commonly available carbon film resistors & Mallory 150 series metalized polyester film capacitors. Tweaked to a spec I was happy with using features from the '67 & '68 50 watt amps, it has served me well & sounded very good. Over the years however, I often thought about a build using more "period correct" components, so started slowly gathering components.
Now I must say a very big "thank you" & "very much appreciated" to one Mr RVA for his assistance in acquiring a complete set of period correct Iskra resistors, without which this project would still be in the "parts aquisition" stage, so thank you very much Ray, you are a scholar & a gentleman. So, with parts acquired the amp was stripped to thisView attachment 20488
Cleaned up & new ceramic pre amp tube tube sockets installed (to replace micalex ones), the board cleaned up & re turreted & rewired, then installedView attachment 20489
& then populatedView attachment 20490 The C296 mustard caps are mostly Philips Mullards, the exceptions being the 0.0022uf V1B coupling cap, which is an Aussie made Philips Miniwatt, made at the Hendon SA plant, & the two 0.68uf/160V V1B & V2A cathode bypass caps, which are Spanish Philips Bianchi types. These "foreign" mustards are made on the same machine types as the Mullards, with the same material stock supplied by parent company Philips, so while never used in Marshall's, they're the same cap type. Not what you'd use in a vintage Plexi restoration but plenty good for clone amps. The HT voltage dropping resistors are period correct 1 watt RS morganites. The treble & treble peaking caps are 60's era RBS 2% 560pf silver mica's. I'm still looking for two RS types for these positions. Volume 1 bright cap is a 4700pf/1000V Lemco ceramic disk, though this may well be changed to a smaller value.
After going through all checks I installed the same valves that had previously been in it, power tubes biased to 66%.
So how does it compare? Well it's noticeably different & to my ears for the better. Much smoother, mids are more open. Seems richer with better harmonic content & has that "note bloom" thing happening much more than it did before. After putting about an hour on it I changed out the tubes for the new ones I had for. Rectifier is a NOS/NIB Valvo branded f32 Blackburn Mullards GZ34. EL34's are the same (St Petersburg) Winged C's that have been in it. Of the two NOS Brimar CV4004's that I had for it, the first that I tried in the phase inverter position gave a plate differential voltage of close to 11v, so it was relegated to the V2 spot. The other Brimar in the phase inverter socket gives a plate differential voltage of 7.8v, a little higher than I'd like but still in the ballpark & will do for now. V1 is a NOS Mullard 163 series 12AX7. I've put around 5 hours on the amp with these tubes in & really like the grit of the 163 Mullards in V1, however it is just a little noisey (not microphone, just a tad noisey on the bright channel). I do have a NOS/NIB factory matched duet of Mullard XF2 EL34's & another 163 series 12AX7 on the way so plan on doing some tube rolling & maybe a couple of small circuit tweaks as the amp settles in & will then try to get a couple of clips up. A couple more picsView attachment 20492
View attachment 20491
Cheers

Better than Bletchley. Good form!
 
Frickin beautiful, Ivan... A work of art! Excuse me for a bit... I'm getting teary eyed :cry:.

You must have a good chunk of dough invested into this. But after a meticulously built affair, and a seemingly successful launch, I bet Mr. James Charles Marshall is smiling from above... :yesway:

Can you share some info on the...
Chassis?
Trannies?
Pots?
 
Yes, do tell....
Awesome, impressive clean build Ivan!
I am no pro but one of the things I am doing moving forward is to keep the filaments wire twist as close as possible to the socket pins, like on the right in the photo.
Unless your heaters are DC.

heater10.jpg
 
Thanks for the kind words guys. Firstly, a couple of points I failed to mention. The V1a cathode arrangement is 820r/ 10uf rather than 820r/330uf. This still gives a "flat" frequency response while eliminating the flubby bass that is usually associated with the "normal" channel 2 on these type amps.
The V1B cathode arrangement is the '68 spec (first year with split cathode) of 820r/0.68uf rather than the common 2k7/0.68uf. This biases the valve a good bit hotter than the more common arrangement does, & while it provides good "grind" it doesn't clean up as well.
One planned tweak is to add a 4k7 (Piher) resistor across the outter lugs of the presence pot. This is something that Marshall occasionally did on some amps. This reduces the value of the phase inverters "tail" resistance from 15k to about 12k4. This makes for a slightly gainier power amp with a little quicker transition into overdrive. Coupled with a tube rectifier & all 33uf filtering this provides a power amp section that I'm partial to.
Chilli, yes, the PT has primary taps for 240, 230, 220, 120 & 100 VAC supply. Our Aussie mains are 240VAC, either well or not so well regulated depending on ones location. I'm not too far from a substation so our supply is pretty steady.
Sysco, all iron is Heyboer. The PT is actually a Marstran/Heyboer 1202-55 JTM45 item. This supplies the same HT voltage & current as the 1202-133 transformer that was used in some '67 & early '68 amps. Though the 6.3VAC heater winding doesn't quite have the same current capability as the -133, Brian assured me it was up to supplying the EL34's & it has proved to be & doesn't run hot.
Jtcnj, I recognise the heater wiring pics from the Valve Wizard site, & have used similar wiring techniques with multiple cascaded gain stage type amp builds. I stuck with the way Marshall did it with this amp for a couple of reasons. One being, if you look at the ceramic 9 pin Noval sockets you will see that in the center, between the pins there is a metal cylindrical peice crimped into the socket. This is there for two reasons. Though the tubes we use do not have it, some tubes have a small metal disk at the bottom of the tube, inside the glass envelope between the pins & attached to the cathode. When the tube is plugged into the socket the metal disk & the "metal cylinder" on the socket form a capacitor. The metal cylinder is tied to ground, thus removing unwanted frequencies (heater hum etc). The 2nd reason the metal cylinder is crimped into the base is that it is holding the upper & lower halves of the socket together. Without it the socket falls apart & the pins fall out. The metal cylinder is quite tall, as tall as the pins, see here15459578918031887899758849604154.jpg
Sorry for the unfocused pic. The micalex socket, which is the type showing the wiring across the socket, does not have this central "metal cylinder"15459581311728339863313481094095.jpg
You can see why wiring a ceramic socket as suggested may not be ideal. If you think about it, where any connection to the valve is at its closest to the heater wiring is right at the socket pins & valve pins, which of course can't be twisted. No denying that twisting the heater wiring is good practise, but where heater supply hum becomes a problem I find that simply elevating the heater supply is adequately effective.
Well, everything's loaded into the wagon, I'm off to make some noise. Cheers

Edit; Sysco, the pots are CTS, but a few were opened up & had the carbon element very carefully shaved to bring the value up to where I wanted, eg, volume pots to close to 1.1meg, treble to almost 300k, middle to 30k. Cheers
 
Last edited:
Sysco, the pots are CTS, but a few were opened up & had the carbon element very carefully shaved to bring the value up to where I wanted, eg, volume pots to close to 1.1meg, treble to almost 300k, middle to 30k.
Those are very small increases in those potentiometer values. What do you like about it?

Regarding the Mid pot: Why not use an off-the-shelf 50K pot?

Do you have any experience or opinions about PEC pots?
 
Those are very small increases in those potentiometer values. What do you like about it?

Regarding the Mid pot: Why not use an off-the-shelf 50K pot?

Do you have any experience or opinions about PEC pots?
Yes, they are only small increases in value. I've found many CTS (& some other brands) pots will meter lower than their stated value, whereas actual Plexi pots, when measured have been found to be a little higher than their stated value, whether through years of use or being originally manufactured like I don't know. So these (which metered low) had their values adjusted some time ago. The "adjustments" made a slight audible difference, but a few slight differences add up. (You don't know where there's any NOS Hunts/Daly or LCR electrolytics do you?)
I haven't used PEC pots, but by all accounts they are very good quality items.
Thanks 67plexi. I prefer the 1202-55 PT too. B+ is 462v with a wall supply of 237VAC (I need to invest in a Furman). Cheers
 
Back
Top