Su-101

An Alternative Layout

SPG designs hit a dead end in many nations by the end of WWII. The concept of a heavier gun on the same chassis stopped working. Regular guns were getting so large that there were issues with the chassis. The results were essentially the same as just putting the gun in a tank with a rotating turret. This was especially true for medium SPGs. Germany, the USSR, and Great Britain eventually ended up with medium tanks that had the same gun as medium SPGs. Several nations (especially Germany and the USSR) ended up overloading the front suspension. This led to a search for new solutions. This led to the Uralmash-1, the most unusual Soviet late war SPG.

In search of a future-proof layout

The SU-100 SPG was accepted into service with the Red Army on July 3rd, 1944. In addition to a more powerful 100 mm D-10S gun, this vehicle received 75 mm of front armour and a commander’s cupola. It became clear that the limit of the chassis was reached. The SU-100 weighed 2 tons more than the SU-85, all of which weighed heavily on the front of the vehicle. Complaints of broken wheels and suspension springs began coming in starting with late 1944. This was not the only issue. The SU-85’s gun had an overhang of under 2 meters, but the SU-100’s gun stuck out by 3350 mm. This resulted in additional issues in combat. Risk of damaging the gun when driving on bumpy terrain, in forests, and in cities increased. Furthermore, the front suspension springs began to sag. This was noticed even on the prototype.

A draft of the SU-100M1, a potential replacement for the SU-100, dated October 7th, 1944.

This was not news to the UZTM design bureau. Work on the SU-D25, a SU-85 with a commander’s cupola and the 122 mm D-25S gun, began in the fall of 1943. The gun’s overhang was 3300 mm and the mass rose to 32 tons. The SU-D25 project evolved into another vehicle, the SU-100P. The issues of heavy load on the front wheels and significant gun overhang were obvious and noted even earlier during work on experimental SPGs on the ISU-152 chassis. The first solutions of these issues also had to do with the heavy SPGs. In late March of 1944 factory #100’s design bureau presented a deeply modernized IS-2 tank with a rear fighting compartment designed under the direction of N.F. Shashmurin. According to the drafts, an SPG with a rear fighting compartment was planned on the same chassis. The work did not proceed for various reasons, and so the ball was in UZTM’s court.

The driver’s compartment was separated from the fighting compartment. Unlike in later projects, the SU-100M1 did not have a passageway connecting the two.

A rear fighting compartment was not a new thing for Soviet SPGs. This layout was first suggested for the A-46 medium tank destroyer back in 1941. It had obvious advantages and disadvantages. One advantage was that the gun overhang was either minimized or disappeared entirely. In case of open or semi-enclosed fighting compartments, it was also much easier to serve additional ammunition during firing.

There were also many disadvantages. The driver’s compartment would be cramped, the engine compartment would have to be more compact, and the dense layout would introduce issues with access to the engine and cooling system. The biggest issue was that a redesign of the chassis would be required. Few tank building nations could allow this during a war. The USSR considered altering the chassis undesirable, as it would require more time to set up production. The SU-76M with its rear fighting compartment is an exception, but it was a necessary evil. The T-70B chassis in its initial form was unsuitable for building an SPG.

The transmission remained in the rear, which resulted in a height increase of 100 mm compared to the SU-100.

Another interesting point is that several nations began to search for a new layout at the same time. This was especially true for Germany. It was clear that the reserves of light and medium chassis were exhausted. The maximum acceptable load on the road wheels was reached and then surpassed, with obvious results. Given that there was already a successful experience with the Panzerjager Tiger(P), Krupp proposed a series of light and medium tank destroyers with a rear fighting compartment in October-November of 1944. The 6th Department of the Ordnance Directorate rejected them all, chiefly due to excess mass. Nevertheless, it’s interesting that designers in different nations had the same ideas.

ESU-100, a variant with an electromechanical transmission.

It is not known when exactly the UZTM design bureau began working on alternative variants of the SU-100. One can confidently say that investigation into the vehicles was already mentioned in the plan for experimental work for the second half of 1944. The document, dated August 22nd, 1944, describes “an SPG on the T-34 and T-44 chassis with a D-10S or D-25 gun and rear fighting compartment”. 100,000 rubles were allocated for the design and blueprints. The due date was Q3-Q4 1944. In case of the T-34 chassis, two vehicles were added to the prospective SU-122P: the SU-100M1 and ESU-100. They were similar in terms of armament, hull layout, and overall concept.

The electromechanical transmission was not just larger, but also heavier. Installing it required a new layout for the driver’s compartment and engine compartment.

The vehicle was only 40 mm longer than the SU-100, but the height of the casemate increased by 100 mm to 2300 mm. This was necessary due to a decreased fighting compartment height in the new layout. While the engine and cooling system moved to the middle of the hull, the gearbox and other transmission elements remained where they were. This required adding a driveshaft leading to the gearbox. A large access panel was installed in the rear of the vehicle to work on the gearbox. This was not a new problem, the experimental GAZ-71 SPG also had a similar layout. Even though the fighting compartment was smaller, the ammunition capacity was greater: 39 100 mm rounds. The mass increased by a ton compared to the SU-100, reaching 32.6 tons. This was not only due to the layout, as the front armour was thickened to 90 mm. The top speed was estimated at 47.5 kph.

The driver had to be shifted to the left.

The driver’s compartment was also altered. Like on the Ferdinand, there was no access to it from the fighting compartment. A lack of space also meant that the fuel tanks were next to the driver. Like on the SU-100, the driver’s hatch was placed in the upper front plate.

SU-100M2, an alternative on the T-44A chassis.

The ESU-100 was similar to the SU-100M1, but it was even heavier: 35 tons. The ESU-100 used an electromechanical transmission developed at factory #627. The transmission was a product of EKV an IS-6 tank (Object 253) development. The transmission was not just heavier, but also bigger, leading to a decrease in ammunition capacity to 35 rounds. Changes were also introduced in the engine compartment. The GT-627 generator forced the designers to shift the engine forward, partially entering the driver’s compartment. The driver was moved to the left, as were the fuel tanks. Due to increased mass, the top speed dropped to 42 kph.

The lighter and lower chassis made the SU-100M2 preferable to the SU-100M1.

As mentioned above, the T-34 was not the only prospective chassis for the SU-100’s replacement. GKO decree #6209s “On organizing T-44 tank production at factories #75 and #264” was signed on July 18th, 1944. The T-44A (an improved variant of the tank) entered trials on August 18th, and on September 23rd NKTP order #573s listed a number of corrections that would have to be made before the tank entered production. It was clear that mass production was around the corner, and designing the SU-100M2 SPG was a logical step.

Sloping of the front hull allowed for a decrease in mass.

The layout of the hull was largely similar to the SU-100M1, and so was the ammunition capacity. However, the new chassis allowed the height to decrease to 2200 mm, the same as the SU-100. The mass was also the same as the SU-100: 31.6 tons. Like on the SU-100M1, the driver sat in the center of his compartment. The slope of the sides was even steeper. The driver’s hatch vanished from the front of the hull and moved to the roof. The thickness of the armour was also increased: the upper and lower front hull were 90 mm thick and the sides were up to 75 mm thick. The speed was estimated to be the same as for the SU-100M1, but the fuel capacity decreased from 560L to 450 L. The tank used a more powerful V-2-44 engine that consumed more fuel.

SU-122-44, an alternative vehicle with a front fighting compartment. This vehicle had highest priority at one point.

The SU-122-44 was the last of the alternatives. This vehicle was a counterpart to the SU-100P. It was developed using the classical layout with a front fighting compartment. This design was a backup plan in case the rear fighting compartment idea did not pan out. One advantage of the SU-122-44 was that it did not require as many changes to the T-44A chassis. However, the barrel overhang was 3070 mm, while it was no more than a meter on its competitors. Thanks to a different layout, thicker armour, and the more powerful D-25T gun, the mass of the SU-122-44 reached 32.8 tons. However, a comparison with the SU-122P shows that the increase was worth it. The SU-122-44 was the roomiest, which not only improved the crew’s working conditions, but also allowed the increase of ammunition capacity to 40 rounds, a third more than the ISU-122. The vehicle also had superior armour, as its 90 mm upper front plate was positioned at an angle of 60 degrees. Based on the performance of the IS-2, this armour protected from the 88 mm Pak 43 gun at a range of 500 meters. The crew of four was an issue. Practice showed that the A-19 and D-25 needed two loaders, otherwise the average rate of fire was 2.5-2.75 RPM. This was confirmed in trials of the SU-122P.

There can be only one

The UZTM design bureau finalized all projects on October 7th, 1944. In all cases the overall direction was given by L.I. Gorlitskiy and the lead engineer was N.V. Kurin. Documentation on all five vehicles including the SU-122P was sent to the NKTP in mid-October. By that point the SU-122P was already built and undergoing factory trials. Trials showed that the average speed of the SU-122P was lower than the SU-100. It was becoming more and more clear that the T-34’s chassis was reaching its limits.

Initial layout of the SU-101 prepared by early March of 1945. The prototype followed these blueprints.

The NTKP’s decision regarding the UZTM’s projects was reasonable given this information. The T-34 chassis was no longer satisfactory, and so the SU-100M1 and ESU-100 were left behind. On October 26th, 1944, Malyshev signed NKTP order #625s choosing the projects on the T-44A chassis. UZTM’s design bureau had to complete documentation for the SU-122-44 and SU-100M2 by December 25th. The SU-122-44 split into two vehicles: the SU44-100 and SU44-122. A variant with the 100 mm D-10S gun would be developed in addition to the one with a D-25S. The chassis was also going to change, in part due to thicker armour. The front armour grew to 120 mm and the sides to 75 mm. This protected the tank from the 88 mm Pak 43 L/71 at any range. Most attention was paid to improving the SU-122-44, presumably the NTKP considered this vehicle the most promising.

The layout changed significantly compared to the SU-100M2, and the armour was radically improved.

Despite high interest in the SU-122-44, the SU-100M2 had its share of required changes. A reworked project was due by January 1st, 1945, where the engine was shifted to the right and the driver was shifted to the left to allow him a passage into the fighting compartment like on the SU-76M. A hatch was placed above the driver like on the T-44 to allow him to drive while sticking out of the hatch. A DShK AA machine gun was installed on the commander’s cupola. A fume extractor was going to be installed on the gun and MDSh smoke bombs were to be carried on the rear. The list of improvements numbered 11 in all.

The sides were sloped at 45 degrees to make them stronger.

The optimism regarding the SU-122-44 was short lived. Calculations for improved armour made in December of 1944 proved that the vehicle was a dead end. The mass of the vehicle grew, but the T-44 showed issues with peeling road wheel tires as is. Trials of the SU-122P also played a part. The T-44, which was supposed to be used without much modifications, was in question too. Factory #75 had difficulty getting production off the ground, and the blueprints sent by factory #183 were already changed. The UZTM was in danger of getting involved with a lengthy battle in mastering production themselves. In short, there was no longer any sense in producing the SU-122-44.

The SU-102. Initially it only had different armament from the SU-101, but in practice the prototype had many internal changes.

UZTM director B.G. Muzrukov declined to work on the SU44-122 and SU44-100, focusing instead on the SU-100M2. Muzrukov informed Malyshev of this decision, who responded with understanding. The GBTU’s SPG branch did not entertain the idea of a “classical” SPG on the T-44 chassis from the very beginning. The 3rd Department of the SPG branch lists “control over the development of an SPG with a rear fighting compartment” as work for the fall-winter of 1944. The SU-122-44 is absent from these plans. A special commission from the NKTP led by I.S. Ber arrived in January of 1945 to review the new blueprints. The commission agreed with the UZTM’s findings. The variant with a rear fighting compartment was more promising. Calculations showed that thicker armour will result in a small increase in mass. Furthermore, the fighting compartment could fit not only the 100 mm D-10S, but the larger 122 mm D-25S. The commission pointed out a series of issues that would have to be corrected before the final project was submitted.

SU-101 hull prepared for testing.

The final version of the SU-100M2 was presented in March of 1945. The mass increased to 32.4 tons due to improvements and changes to the armour. The NKTP approved this submission with NKTP decree #107s signed on March 7th, 1945, which also gave the vehicle the name “Uralmash-1”. According to the order, the prototype armed with the 100 mm D-10S gun was due on May 1st. It was named SU-101. A second prototype with the 122 mm D-25S gun was due on May 15th. This vehicle was named SU-102. An Uralmash-1 hull was also expected for penetration trials. Some final changes were introduced before production. The cooling system and fighting compartment ventilation system would be altered, the engine compartment was insulated, the ammunition capacity was increased to 40 rounds, the suspension was reinforced, and protection of the gun and lower front plate was improved.

1 Like

The hulls of the SU-101 and SU-102 produced at the UZTM in March of 1945 were the same.

The SPG branch of the GBTU issued 1,500,000 rubles for design and production of two Uralmash-1 prototypes (750,000 each) in Q2 of 1945, but work on the project already began in March. The UZTM’s hull workshop began building three hulls as specified in order #107s. Documentation on the Uralmash-1 hull was therefore finished in March.

Changes made to the T-44 chassis by factory #75 in late 1944 were included during production.

The final variant of the Uralmash-1 had many changes compared to the SU-100M2. For instance, the front armour of the casemate was thickened to 120 mm and the sides to 90 mm. The sides were also heavily sloped, to 45 degrees. This made the casemate more resistant to flanking fire. Attention to protecting the sides was no accident. Operational research showed that Soviet vehicles were often hit from the sides. Unlike German tanks and SPGs, new Soviet AFVs had heavily protected sides. This led to some trouble with placing the crew, for instance the commander had to get a special protrusion for his cupola. The number of hatches in the roof decreased compared to the SU-100M2. Only the commander’s cupola and panoramic sight hatches remained. The ventilation fan was moved from the center to the right rear corner.

Unlike the front and sides, the rear of the casemate was rather thin. This showed in trials.

The UZTM received orders from the NKTP and GBTU on April 28th, 1945, instructing it to perform penetration trials. The trials program arrived two days later. It required the hull to be fired on by the 85 mm gun, 100 mm D-10S, and German 88 mm Pak 43 L/41. The guns fired HE as well as AP. Trials began with firing 88 mm HE shells. Strikes against the upper glacis damaged the bolts holding the gun mantlet. Firing at the sides led to worse damage as the weld seams connecting the sides and floor cracked. Two more hits on the left side broke the seam holding the commander’s cupola visor and the cupola itself. Similar damage resulted when firing at the right side. Firing at the rear led to greater damage: it destroyed the right plate and tore out the exit hatch.

The front of the casemate was impenetrable to 85-100 mm rounds. However, the shockwave broke off the gun mantlet retaining bolts.

The hull was hit 11 times with 88 mm HE. The hull design was deemed acceptable despite the damage when firing on the sides and rear. Two variants of assembly were proposed to strengthen the sides, both of which included interlocking plates.

An unfortunate consequence of hitting the side with an HE shell. The seam holding the commander’s cupola burst.

The largest proportion of the trials (52 shots) fell to German 88 mm AP rounds. It turned out that the upper front plate of the hull and the front of the casemate were impervious to attack at all ranges. The lower front plate could be penetrated at 740 meters. The sides were worse. Calculations showed that the critical range was 867 meters, and the left side was more vulnerable due to the commander’s cupola bulge and layering of the metal. As for the vertical 75 mm thick sides, they were not a serious obstacle for the German gun at any range.

The upper front plate could not be penetrated.

Soviet 85 and 100 mm shells had a similar effect. They were not tested as heavily. Only 3 shots were made from the D-10S and 22 shots from an 85 mm gun. As with the Pak 43, the upper front plate and front of the casemate were not penetrated. The overall results showed that the hull was sufficiently robust, but there was food for thought. For one, there were issues with the quality of welding (a typical defect at UZTM). There were also issues with the joints between the sides and floor of the hull, as well as the commander’s cupola visor which was vulnerable to both AP and HE shells. There were also questions about the design of the exhaust manifolds.

Results of hitting the rear of the hull.

The commission commented on the need to protect the lower front plate and reinforce the commander’s cupola visor. Separate trials of a fully assembled gun mount were also needed. The rear of the hull was unacceptable and needed to have equivalent resistance to the SU-100. The exhaust manifolds and plate joints also had to be strengthened.

Proposed strengthened plate joints.

NII-48 representative G.I. Fedoseenko had his own opinion. He considered that there was no need to work on protecting the floor, as the odds of it being hit were very low. The issue of optimal armour composition was to be resolved jointly with the NII-48. The NII-48 report regarding the SU-101 noted that the UZTM design bureau created a very good hull and casemate design. It was significantly more robust than previously seen heavy and medium SPG casemates.

Heated armoured slipper

UZTM eagerly set out to complete the task set for them by the NKTP and GBTU. Three Uralmash-1 hulls were finished in March of 1945, and the first prototype was delivered for trials on April 27th, even before the deadline. As ordered, the first vehicle was the SU-101 armed with the 100 mm D-10S gun.

SU-101 prototype on a test run, late April of 1945.

The project changed a lot since the SU-100M2, and this included the layout. The idea of a rear transmission was discarded, as it made the fighting compartment smaller and the transmission more complicated. Now the transmission and drive wheels were on the front. This made service more difficult, but this was not critical. The engine deck was composed of two halves. To remove the gearbox, one would turn the gun all the way to the left, lift the front half of the engine deck, remove the radiator, and then access the gearbox. This was complicated, but not as complicated as the acrobatics necessary to take out the gearbox of a Jagdpanther. Removing the engine was harder, this was impossible without taking out the gun.

Maximum gun elevation.

The fighting compartment was reworked. The size of the compartment decreased due to the requirement for greater protection, this meant that the ammunition capacity decreased from 40 rounds to 36. At the same time, the height of the fighting compartment increased due to the disappearance of the transmission and driveshaft. The reduction in ammunition storage for the cannon was also affected by the introduction of the DShKM AA gun. 7 boxes of ammunition were carried in the fighting compartment in addition to 16 magazines for the PPSh.

Maximum gun depression.

The reworked layout and thicker armour had an impact on the mass of the SU-101. It reached 34.09 tons, 1.5 tons more than the proposal made in March of 1945. The rear fighting compartment allowed for the mass to be more evenly distributed along the chassis. The gun overhang was also only 600 mm. At the same time, there were some requirements that were not completed due to the rush. For instance, the passageway from the driver’s compartment to the fighting compartment was present, but the engine compartment was not insulated. There were also other issues that came up during factory trials.

The SU-101 had the best protection of any SPG in the world.

A trial run of the SU-101 held on April 27th, 1945, gave some preliminary results. The trials showed that the weight was distributed evenly and the torsion bar suspension worked well. Due to the more powerful V-44 engine the mobility of the SPG remained on par with the SU-100. Unlike on the T-44, the driver’s hatch did not rotate, but he could still look out of it while driving or use the MK-IV rotating periscope. There was also praise for the large hatch in the rear that made it easy to enter the vehicle and load ammunition.

Spare track links improved frontal protection at large and medium distances.

There were also plenty of drawbacks. One of the biggest was overheating of the driver’s compartment due to a lack of insulation of the engine compartment. One may recall the T-60, T-70, and SU-76M where the engine bulkhead was largely absent, but this was a medium SPG with a much more powerful engine and not a light vehicle. Heat also came from the gearbox and brake bands. The driver suffered from serious overheating in the spring. Even with an open hatch he could drive for only 7 km on a highway before he had to exit the vehicle. The water and oil temperature reached 100-110 degrees. The plan to drive for 10-15 km had to be abandoned.

The rear hatch made it easy to enter the vehicle and to load ammunition.

There were issues other than overheating. The gearbox lever was too far to the right and also heated up. The idea of attaching the control levers to the upper front plate was criticized. There was a chance of them being put out of action after a hit to the upper front plate, plus they were uncomfortable to use. The location of the levers was not considered when the SU-101 hull was shot up in trials.

A visual comparison between the SU-101 and SU-100. The former appears larger, but this is a trick of perspective.

Issues that arose during the first factory trials led to significant changes in the second vehicle. The SU-102 was late as a result, arriving only on June 24th, 1945. Initially the only difference from the SU-101 was the 122 mm D-25S gun. Thanks to the heavier cannon, the mass of the SPG rose to 34.77 tons and the ammunition capacity was reduced to 28 rounds. The SU-122P carried only 26 rounds, and that’s not including the 4 boxes of ammunition for the DShKM.

The final SU-102, NIBT proving grounds, early 1946.

10 changes were made to the SU-102 overall. The controls were altered. Most of the levers were moved from the upper front plate to the floor. The driver’s seat was moved 250 mm back and new observation devices were installed. A lot of work was done to prevent overheating. Additional cooling fans for the gearbox, final drives, and engine were introduced. Bulkheads were also introduced that covered transmission elements. The water radiator was enlarged.

Various changes were made to the vehicle during trials.

The test run showed that it was no more difficult to drive the SU-102 than the SU-100. The situation with the temperature improved, at least now the vehicle could travel for 25-30 km without stopping. The overheating persisted at low gears, which UZTM admitted to. Both vehicles were sent back to be improved and were subjected to modernization before returning to factory trials. The air for cooling the engine was now taken from the fighting compartment rather than the engine compartment. Double bulkheads with an asbestos liner were introduced to protect the driver’s compartment. The bulkheads were also airtight.

The SU-102 approached a weight of 35 tons, which affected reliability.

Both vehicles visited the shooting range before mobility trials. The SU-101 made 10 shots in 2 minutes 15 seconds, using both the left and right ammunition racks. The concentration of fumes in the fighting compartment was normal, the electric firing mechanism worked well. The SU-102 also performed well, showing a rate of fire of 2 RPM.

SU-101 fording a river, July 1945.

Mobility trials began on July 24th, 1945. The plan was to drive 300 km to to village of Ognevo and back. The SU-102 was quickly disqualified, as the gearbox broke after 58 km. The SU-101 drove between July 24th and 27th, covering 210 km on a highway and 92 km on bad country roads. A top speed of 53.8 kph was reached. The SU-102’s top speed was 48.2 kph, and at this speed it showed issues with track pins. The SU-101’s average speed was 18.5 kph and fuel expenditure reached 198-219 L per 100 km. Water temperature reached 112 degrees and oil temperature was 108 degrees. The temperature inside the driver’s compartment was more interesting: 33 degrees with an open hatch and 66 degrees with a closed hatch. Recall that the SU-101 had issues with overheating from the very beginning.

Trials showed that the vehicle could cross a 0.9 meter deep river.

The vehicle took part in fording and climbing trials. The SU-101 easily crossed a 0.9 meter deep river, 23 degree hill, and tilt at the same angle. More serious obstacles could not be found. Testers indicated that the SU-101 had a reserve of power. The trials were not completed either way as the diffuser liner broke at 255 km.

Driving on a slope.

The SU-101 was submitted for repairs on July 29th, which revealed many defects. On August 4th the UZTM director issued order #30s, which introduced 12 changes to the SU-101’s design. A large potion of them was linked to the cooling system. The vehicle returned to trials on September 25th. It drove 106 km on a highway and 9 km on country roads over the course of two days. Progress was made with cooling: the temperature in the driver’s compartment did not exceed 45 degrees and was usually around 24-33 degrees (at an ambient temperature of 6 degrees). The control rod for the left final drive clutch burst after driving for 113 km and the SU-101 was towed back to the factory. Road wheel tire damage characteristic of the T-44 began to appear. Nevertheless, UZTM managed to deal with the main problem: overheating.

The 23 degree slope was conquered without much effort.

The GBTU and GAU continued brainstorming while the Uralmash-1 was being improved in Sverdlovsk. The end of the Great Patriotic War did not mean that the arms race came to an end, if anything the situation was the opposite. The GAU composed a list of work for 1946 in October of 1945. It included the development and production of a “100 mm high power gun for a medium SPG with a rear fighting compartment”. The SPG branch of the GBTU developed its plans for 1946-1950 even earlier, in late August. They called for an open topped SPG on the Uralmash-1 chassis armed with either a 122 mm D-25S gun or 152 mm ML-20 gun-howitzer. The plans also included a SPAAG with two 57 mm guns and a mobile observation post.

The SU-102 during trials at the NIBT Proving Grounds. By then the Uralmash-1 was not very interesting to the GBTU.

None of these plans ever came to pass. A number of issues experienced by the Uralmash-1 vehicles were caused by an overloaded chassis. They could be resolved only by switching to the more reliable T-54 chassis. This was proven by trials of the SU-101 held in 1946. The NIBT Proving Grounds deemed the vehicle too cramped and too hot. There was also no one left in Sverdlovsk to keep working on this vehicle. The design bureau was rapidly downsizing, production of the SU-100 ceased, and UZTM transitioned to peacetime production. There were no longer any work on the Uralmash-1 planned for 1946. The work began from scratch at factory #174 in Omsk, giving the Object 600 that was accepted into service with the Soviet Army on March 15th, 1954. As for the Uralmash-1, it was a dead end in Soviet medium SPG development. The SU-101 survives to this day and can be seen at Patriot Park.

1 Like